The future of vegan cheese is being crafted by Climax Foods, a Berkeley, CA startup founded by Dr. Oliver Zahn, a scientist who previously worked for forward-thinking companies GoogleX, SpaceX, and Impossible Foods. 

The startup has thus far raised $27 million to fund its mission of recreating the world’s most desired animal foods by unlocking what it calls a “smarter path from plants to food.” This path is paved by technology supported by two years of rigorous research that resulted in a “deep plant intelligence” platform Climax is using to bring its first products to market. 

Climax Foods

On the horizon are vegan “moonshot” cheeses—or products that are intended to be disruptive and future-oriented—in four varieties: Blue, Brie, Feta, and Chèvre. Cultured and aged, these cheeses are made with a combination of seeds, legumes, and plant oils, avoiding allergens and GMOs to appeal to a broader consumer base. These cholesterol-free vegan cheeses were also developed to compete with traditional dairy cheeses on taste, nutrition, and price—the three main drivers of purchases.  

“We see a bright future as the foundational work that Climax has done to unlock the full flavor potential of plants will open countless doorways to an exciting planet-supporting food industry,” Tom Chi of At One Ventures, a co-founder of GoogleX and one of the company’s early investors, said in a statement.

How deep tech can improve vegan cheese

Can plants really make better cheese than cows? Zahn thinks so. Born in Bavaria, Germany, Zahn’s work is informed by his understanding of the intricacies of space and time and motivated by a desire to affect positive change on a global scale.

Climax Foods

“We started from a profound appreciation for the complex flavors and textures of dairy products,” Zahn said in a statement. “Cows have made our milk for thousands of years. However, less than 10 percent of the plants they eat get turned into food for humans, which has led to significant environmental and health problems in today’s much more crowded world.” 

Working inside a converted chocolate factory in Berkeley, CA, Zahn and his team of 40 scientists began developing their vegan cheeses inside based on a deep understanding of their dairy counterparts on a molecular level. To best approximate dairy cheeses, Climax relies on its machine-learning frameworks to identify the ideal building blocks for its recipes using the world’s 300,000 edible known plants. 

“It is human nature to rethink ancient practices, so we came up with a smarter way,” he said. “By using data science to accelerate plant-based ingredient and process discoveries, we are saving thousands of years of tinkering to create products that are just as tasty as the cow-based predecessors without the downsides, today.”

The startup recently served its vegan cheeses at tasting events in New York City, Los Angeles, and San Francisco. Currently, Climax is expanding beyond its Berkeley facility with a new cheese production plant in Petaluma—led by cheese industry veteran Caroline Di Gusto—and plans to debut its first products to the public through soon-to-be-announced distribution partners in the coming months.  

Climax Foods

“As the company’s first end-to-end production facilities, these locations will allow us to significantly increase the speed and scale of our production and distribution,” Di Gusto said in a statement. “Throughout the process, we’ve sought out ways to improve the company’s cost structure and sustainability of operations. We are set to produce plant-based cheeses at price parity with dairy for North American distribution.”

In the future, Climax will focus on additional “moonshot” categories that can be improved with its animal-free approach.

Climax Foods

“Our technology and ingredient discoveries will soon power the replacements of bigger categories with successors that will be equally delicious and nutritious but more sustainable and—because our products are not heavily processed—substantially more economical and environmentally friendly,” Zahn said. 

The cow-free future of cheese 

Globally, dairy cheese is valued at $800 billion and producing it at this large scale comes with a number of downsides, including contributing to environmental degradation and animal exploitation. Luckily, in addition to Climax, a growing number of companies are working to improve dairy cheese (and plant-based cheese) using technologies such as artificial intelligence and microbial fermentation. 

While Climax is showcasing its capabilities in the world of artisan cheeses, fellow California-based company New Culture is focusing on commercializing an improved version of mozzarella—America’s most-consumed cheese.

New Culture

New Culture’s approach focuses on microbial fermentation, where microbes take the place of cows in producing real (but animal-free) casein, a functional protein that delivers the desirable stretch and melt of dairy cheeses. While New Culture can make a variety of vegan cheeses using this innovative protein, it is initially focusing on mozzarella to create the biggest impact. 

The company is getting ready to launch its innovative animal-free mozzarella at pizzerias nationwide this year, supported by agribusiness giant ADM and CJ CheilJedang (a $23 billion South Korean company that holds a 25-percent stake in the United States frozen pizza market). 

This post was originally published on VegNews.com.

Highway surveillance footage from Thanksgiving Day shows a Tesla Model S vehicle changing lanes and then abruptly braking in the far-left lane of the San Francisco Bay Bridge, resulting in an eight-vehicle crash. The crash injured nine people, including a 2-year-old child, and blocked traffic on the bridge for over an hour.

The video and new photographs of the crash, which were obtained by The Intercept via a California Public Records Act request, provides the first direct look at what happened, confirming witness accounts of what happened at the time. The driver told police that he had been using Tesla’s new “Full Self-Driving” feature, the report notes, before the Tesla’s “left signal activated” and its “brakes activated,” and it moved into the left lane, “slowing to a stop directly in [the second vehicle’s] path of travel.”

Just hours before the crash, Tesla CEO Elon Musk had triumphantly announced that Tesla’s Full Self-Driving capability was available in North America, congratulating Tesla employees on a “major milestone.” By the end of last year, Tesla had rolled out the feature to over 285,000 people in North America, according to the company.

Tesla Full Self-Driving Beta is now available to anyone in North America who requests it from the car screen, assuming you have bought this option.

Congrats to Tesla Autopilot/AI team on achieving a major milestone!

— Elon Musk (@elonmusk) November 24, 2022

The National Highway Traffic Safety Administration, or NHTSA, has said that it is launching an investigation into the incident. Tesla vehicles using its Autopilot driver assistance system — Full Self-Driving mode has an expanded set of features atop Autopilot — were involved in 273 known crashes from July 2021 to June of last year, according to NHTSA data. Teslas accounted for almost 70 percent of 329 crashes in which advanced driver assistance systems were involved, as well as a majority of fatalities and serious injuries associated with them, the data shows. Since 2016, the federal agency has investigated a total of 35 crashes in which Tesla’s “Full Self-Driving” or “Autopilot” systems were likely in use. Together, these accidents have killed 19 people.

In recent months, a surge of reports have emerged in which Tesla drivers complained of sudden “phantom braking,” causing the vehicle to slam on its brakes at high speeds. More than 100 such complaints were filed with NHTSA in a three-month period, according to the Washington Post.

The child injured in the crash was a 2-year-old who suffered an abrasion to the rear left side of his head as well as a bruise, according to the incident detail report obtained by The Intercept. In one photograph of the crash, a stroller is parked in front of the car in which the child was injured.

As traditional car manufacturers enter the electric vehicle market, Tesla is increasingly under pressure to differentiate itself. Last year, Musk said that full self-driving was an “essential” feature for Tesla to develop, going as far as saying, “It’s really the difference between Tesla being worth a lot of money or worth basically zero.”

The term “Full Self-Driving” has been criticized by other manufacturers and industry groups as misleading and even dangerous. Last year, the autonomous driving technology company Waymo, owned by Google’s parent company, announced that it would no longer be using the term.

“Unfortunately, we see that some automakers use the term ‘self-driving’ in an inaccurate way, giving consumers and the general public a false impression of the capabilities of driver assist (not fully autonomous) technology,” Waymo wrote in a blog post. “That false impression can lead someone to unknowingly take risks (like taking their hands off the steering wheel) that could jeopardize not only their own safety but the safety of people around them.”

Though Waymo doesn’t name any names, the statement was “clearly motivated by Musk’s controversial decision to use the term ‘Full Self Driving,’” according to The Verge.

Along the same lines, the premier lobbying group for self-driving cars recently rebranded from the “Self-Driving Coalition for Safer Streets” to the “Autonomous Vehicle Industry Association.” The change, the industry group said, reflected its “commitment to precision and consistency in how the industry, policymakers, journalists and the public talk about autonomous driving technology.”

Secretary of Transportation Pete Buttigieg has also been critical of the emerging driver assistance technologies, which he stresses have not replaced the need for an alert human driver. “I keep saying this until I’m blue in the face: anything on the market today that you can buy is a driver assistance technology, not a driver replacement technology,” Buttigieg said. “I don’t care what it’s called. We need to make sure that we’re crystal clear about that — even if companies are not.”

Though the language may be evolving, there are still no federal restrictions on the testing of autonomous vehicles on public roads, though states have imposed limits in certain cases. Tesla has not announced any changes to the program or its branding, but the crash was one of multiple that month. Several days prior to the Bay Bridge accident, on November 18 in Ohio, a Tesla Model 3 crashed into a stopped Ohio State Highway Patrol SUV which had its hazard lights flashing. The Tesla is likewise suspected of having been in self-driving mode and is also being investigated by NHTSA.

NHTSA is also investigating a tweet by Musk in which he said that Full Self-Driving users would soon be given the option to turn off reminder notifications for drivers to keep their hands on the steering wheel. “Users with more than 10,000 miles on FSD Beta should be given the option to turn off the steering nag,” a Twitter user posted on New Year’s Eve, tagging Musk.

“Agreed, update coming in Jan,” Musk replied.

Additional reporting by Beth Bourdon.

The post Exclusive: Surveillance Footage of Tesla Crash on SF’s Bay Bridge Hours After Elon Musk Announces “Self-Driving” Feature appeared first on The Intercept.

This post was originally published on The Intercept.

by Peter Elkind

ProPublica is a nonprofit newsroom that investigates abuses of power. Sign up to receive our biggest stories as soon as they’re published.

To many people, the notion that cellphones or cell towers might present a health risk long ago receded into a realm somewhere between trivial concern and conspiracy theory. For decades, the wireless industry has dismissed such ideas as fearmongering, and federal regulators have maintained that cellphones pose no danger. But a growing body of scientific research is raising questions, with the stakes heightened by the ongoing deployment of hundreds of thousands of new transmitters in neighborhoods across America. ProPublica recently examined the issue in detail, finding that the chief government regulator, the Federal Communications Commission, relies on an exposure standard from 1996, when the Motorola StarTAC flip phone was cutting edge, and that the agency brushed aside a lengthy study by a different arm of the federal government that found that cellphone radiation caused rare cancers and DNA damage in lab animals. The newest generation of cellphone technology, known as 5G, remains largely untested.

Here’s what you need to know:

Do cellphones give off radiation?

Yes. Both cellphones and wireless transmitters (which are mounted on towers, street poles and rooftops) send and receive radio-frequency energy, called “nonionizing radiation.” The amount of this radiation absorbed by the human body depends on how close a person is to a phone and a cell transmitter, as well as the strength of the signal the phone needs to connect with a transmitter. Cellphones displaying fewer bars, which means their connection with a transmitter is weak, require stronger power to communicate and so produce more radiation. Wireless transmitters, for their part, emit radiation continually, but little of that is absorbed unless a person is very close to the transmitter.

What does the science say about this? Is it harmful?

That’s the multibillion-dollar question. Government-approved cellphones are required to keep radiation exposure well below levels that the FCC considers dangerous. Those safeguards, however, have not changed since 1996, and they focus exclusively on the unlikely prospect of “thermal” harm: the potential for overheating body tissue, as a microwave oven would. The government guidelines do not address other potential forms of harm.

But a growing body of research has found evidence of health risks even when people are exposed to radiation below the FCC limits. The array of possible harms ranges from effects on fertility and fetal development to associations with cancer. Some studies of people living near cell towers have also confirmed an array of health complaints, including dizziness, nausea, headaches, tinnitus and insomnia, from people identified as having “electromagnetic hypersensitivity.”

The most sensational — and hotly debated — health fear about wireless radiation is cancer. In 2011, the International Agency for Research on Cancer, an arm of the World Health Organization, cited troubling but uncertain evidence in classifying wireless radiation as “possibly carcinogenic to humans.” In 2018, a study by the federal government that was nearly two decades in the making found “clear evidence” that cellphone radiation caused cancer in lab animals. A major study in Italy produced similar results.

Do cellphones pose any special health risks for kids?

Some experts say they do, citing studies suggesting children’s thinner, smaller skulls and developing brains leave them more vulnerable to the effects of cellphone radiation. The American Academy of Pediatrics embraces this concern and has for years urged the FCC to revisit its radiation standards, saying they don’t adequately protect kids. More than 20 foreign governments, as well as the European Environment Agency, urge precautionary steps to limit wireless exposure, especially for children.

What about risks in pregnancy?

A Yale study found hyperactivity and reduced memory in mice exposed to cellphone radiation in the womb, consistent with human epidemiological research showing a rise in behavioral disorders among children who were exposed to cellphones in the womb. Dr. Hugh Taylor, the author of the mouse study and chair of the obstetrics, gynecology and reproductive sciences department at the Yale School of Medicine, told ProPublica: “The evidence is really, really strong now that there is a causal relationship between cellphone radiation exposure and behavior issues in children.”

What does the U.S. government say about cellphone radiation?

The key federal agencies — the FCC and the Food and Drug Administration — have echoed the wireless industry and a number of other groups in rejecting evidence of any “nonthermal” human health risk, saying it remains unproven. The government websites also reject the claim that children face any special risk.

In 2019, during the administration of President Donald Trump, the FCC shut down a six-year review of its 1996 wireless-radiation safety standards. The agency rejected pleas to make the standards more stringent, saying it had seen no evidence its safeguards were “outdated or insufficient to protect human safety.” In 2021, however, a federal appeals court ordered the FCC to revisit the issue, saying the agency had ignored evidence of an array of noncancer harms to humans, animals and the environment, and that its decision to uphold its exposure standard failed to meet “even the low threshold of reasoned analysis.” The FCC has taken no formal action since then.

Why is the issue not resolved?

Determining wireless radiation’s health effects with certainty is difficult. Researchers cannot ethically subject people to endless hours of cellphone radiation to gauge the results. Scientists have to rely on alternatives such as animal studies or epidemiological research, where challenges include getting subjects to accurately recount their wireless use and pinpointing the specific causes of disease or harm. Many health effects of toxic exposure, especially cancer, take years or decades to appear. And the mechanisms of how wireless radiation could affect the body at the cellular level are poorly understood.

Research funding on the issue has also been scarce in the U.S., despite frequent calls for more study. Research (and researchers) raising health concerns have come under sharp attack from industry, and government regulators have remained skeptical. A key FDA official, for example, dismissed the relevance of the federal study that found “clear evidence” of cancer in lab animals, saying it wasn’t designed to test the safety of cellphone use in humans, even though his agency had commissioned the research for that reason.

Linda Birnbaum, who led the federal agency that conducted the cellphone study, said that while proof of harm remains elusive, what is known means that precautions are merited. “Do I see a smoking gun? Not per se,” she told ProPublica. “But do I see smoke? Absolutely. There’s enough data now to say that things can happen. … Protective policy is needed today. We really don’t need more science to know that we should be reducing exposures.”

If I’m concerned about the risk, are there precautions I can take to protect myself and my family?

Because exposure varies dramatically with your proximity to the source of the radiation, experts say a key to minimizing risk is increasing your distance from the phone. This means keeping any cellphone that’s turned on away from direct contact with your body. Don’t keep it in your bra, in your pocket or (especially if you’re pregnant) against your abdomen, they say. And instead of holding the phone against your head when you talk, use a speaker or wired earphones. (Wireless headsets, such as AirPods, also emit some radiation.) Try to avoid making calls when the phone is telling you the signal is weak because that boosts the radiation level. You can also limit exposure by simply reducing how much time you spend talking on your cellphone and texting instead, they say. Using an old-fashioned landline avoids the problem altogether.

Finland’s state-owned research institute center, VTT Technical Research Centre of Finland, has developed a breakthrough technology to crack the code for perfecting plant-based meat. VTT claims that it possesses a “unique toolbox” that combines cellular agriculture and local plant-based ingredients to replicate the building blocks of animal meat.

The research center’s multi-technique approach will allow food producers to turn underutilized, climate-resilient plant crops such as fava beans and sorghum into burgers, steaks, and sausages that offer a real, authentic experience and can be produced on an industrial scale.

“We use cellular-agriculture-based techniques combined with our long-term expertise working with challenging, plant-based raw materials to transform nutritious, local crops into delicious foodstuffs that offer an identical sensory experience to animal-based meat,” Nesli Sözer, Research Professor at VTT, said in a statement. 

VTT

“Our innovation toolkit allows producers to perfectly recreate the sizzling fat and juicy tenderness of meat like never before,” Sözer said. 

A sustainable solution for animal meat

VTT believes its approach is a sustainable, scalable solution to the multitude of environmental and ethical issues of animal meat. Meat is a major source of greenhouse gas emissions, and meat production tends to be overly centralized and vulnerable to supply chain disruptions. 

It is well-known that a new, sustainable food system requires a dietary shift towards plant-based proteins. VTT says that by leveraging this new innovation, food producers can meet soaring global demand for meat while shoring up food security and resilience. 

“In order to solve these challenges, we need plant-based meat that can genuinely satiate demand and that can be scaled in a way that improves food security and nutrition for all,” Sözer explained.  

The VTT researchers know that to have global acceptance of plant-based meat alternatives, they need to match meat’s experience, nutritional profile, and price. “Consumer acceptance has a key role in the development of feasible business around plant-based protein ingredients for meat alternatives. It’s not enough for something to look like meat,” Sözer said. “We need nutritional and healthy foods that are indistinguishable from animal-based meat in every way, that allows people everywhere to enjoy the experience of meat, without the guilt.”

Making vegan meat at a global scale 

Although the meat alternatives industry has grown rapidly in previous years, VTT believes there is still a long journey ahead before plant-based meat can truly gain a foothold in the market. As global demand for meat continues to skyrocket, VTT stands ready to collaborate with food and raw-material producers to accelerate alternative meat innovations and empower them to stay ahead of the curve when it comes to scale, nutrition, and sustainability.  

VTT

Plant-based food, cellular agriculture, and agile manufacturing are key areas that can contribute to providing new opportunities for sustainable food systems. “By combining our experience in plant-based ingredients, food design, and cellular agriculture, we can create proof-of-concept of prototypes for tasty, identical meat alternatives that are ready to be scaled up for industrial production,” Sözer said.

“By supporting companies to develop sustainable production processes and products that have broad customer appeal, we can also work together towards a future that is free from food crises,” Sözer concluded.

Finland invests in plant-based sector

In 2020, Finland invested €2.1 million ($2.3 million) to develop the country’s plant-based sector. The project—led by VTT and the University of Helsinki—has since been working with companies in various areas, from crop producers to plant-based food brands, with the mission of strengthening the position of Finland as a leader in the lucrative plant-based food industry. 

Golden & Green

Previously, researchers at VTT worked on a project that aimed to extract novel plant-based protein from cereals and grains in an effort to reduce food waste. Project leader Emilia Norlund believed that novel sources of plant protein—which the project aimed to promote for use in prepared foods in place of dairy and meat ingredients—extracted from crops typically fed to farmed animals will create a more efficient global food system. 

“Cattle farming is a big issue in climate change,” Nordlund said. “So, if we can move toward using [more] plant-based foods and not feed the crops to animals, there is no doubt that it would benefit the whole planet. Indeed, there are many calculations that show that plant-based food is much more sustainable when we think about the environmental issues.” 

This post was originally published on VegNews.com.

In 2008, the Silicon Valley-focused blog Valleywag published a letter from a “Tesla insider” stating that the company only had about $9 million in cash on hand. Four days later, a Tesla employee apologized for writing the letter. When recently asked on Twitter how Tesla identified the leaker, Musk responded that “we sent what appeared to be identical emails to all, but each was actually coded with either one or two spaces between sentences, forming a binary signature that identified the leaker.”

Curiously, Musk’s recollection of how the Tesla leaker was caught is different from an account provided by Ashlee Vance in his 2015 biography, “Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future.” Vance states that Musk retyped the letter into a Word document, printed it, and then looked through printer logs to find who else had printed a document of the same size. Though a retyped document is unlikely to be byte-for-byte identical to the original letter, given the fluctuations in file size based on metadata and the like, the recreated letter would nonetheless be of comparable size, plausibly giving Musk a ballpark size to look for when auditing printer logs.

But there are yet other accounts of how the leaker was caught. The Sunday Times and Gawker, for instance, both reported that the leak investigation involved taking fingerprints from a printout near a copier, though neither publication explained how the fingerprints were used to identify a leaker. Those accounts raise the curious question of how Tesla or its investigators might have had access to employee fingerprint records.

Regardless of the particular methods used to identify the Tesla leaker, and whether Musk is indulging in a spot of parallel construction, the key takeaway for leakers at Musk’s newest and chaotic company, Twitter, is that they should not print out (or even compose) letters using company resources.

To begin with, a wide array of document watermarking measures can identify the source of a leak. That’s why leakers and publishers need to figure out whether a given document is unique and whether it is safe to publish the document itself — or maybe, in the interest of protecting the source, not publish or even write about the document at all.

The notion of uniquely fingerprinting or watermarking each version of a digital text using various spacing modifications is not particularly new. It has been discussed since at least the early 1990s, with research building on general fingerprinting literature from the early 1980s. Ironically, one of the original proposed applications of document watermarking was to protect newspaper and magazine articles from unauthorized distribution.

Every spatial element of a document — including the spacing between characters, words, sentences, and paragraphs — can be modified in every version to form a unique signature that identifies the recipient of that particular document. For instance, a version of a document sent to one person could have slight variations in the distance between certain characters, words, sentences, or paragraphs that uniquely differentiate the document from a version sent to another person with ever-so-slightly different spacings.

As Musk pointed out, a very primitive spatial watermarking scheme could code a single space after a sentence as a ‘0’, and a double space as a ‘1’, resulting in a “binary signature.” If every copy of an email has a unique spacing pattern, an organization can determine the specific recipient of a leaked email.

One of the original proposed applications of document watermarking was to protect newspaper and magazine articles from unauthorized distribution.

Of course, the amount of possible unique watermarks is dependent on the size of the available text, but that size doesn’t need to be large for the watermarking scheme to be sufficient. In the basic watermarking approach described by Musk, the number of possible unique emails doubles for every added sentence. A two-sentence email could have four unique permutations with both sentences having a double space, both having a single space, or one having a single space and the other a double space, and so on. A nine-sentence email could have up to 512 such permutations — and that would be more than enough to uniquely identify every Tesla staff member as of October 2008, when the company reportedly had under 400 employees. It should likewise be kept in mind that in a more complex spacing watermark scheme, seemingly errant spaces could also be introduced between words or even between characters, under the veneer of being typos, which would greatly increase the number of possible unique permutations even for a modest body of text.

The rub is that if an organization has hundreds or thousands of staff, it would need to create a watermarking (and accompanying distribution) system to match. This system could involve having the sender manually modify each email, or it could be an automated system that creates unique permutations of a given text and keeps track which employee is assigned each permutation.

This leads to a basic check that would-be leakers should apply prior to sharing an email or a document. Was the email sent to an individual email address or to a group email address? If the email was sent to a group address, is this an address that’s been used before, or is the address slightly off, perhaps including a stray character or number?

If an email was sent to an individual address, the chances are higher that it could be watermarked. However, an email can be watermarked even if sent to a group address. For instance, a sophisticated (and hypothetical) system could modify the membership of a group email address to only contain a single recipient for each permutation of an email. The group membership is temporarily modified to remove everyone but a single staff member who is sent a uniquely watermarked version of an email. That staff member is then removed from the group membership and another staffer is added who receives a different version, then that staffer is deleted and another is added who receives yet another version — and on and on the ruse goes until all staff receive an email that appears to have been sent to a group, but which in fact is unique to each staffer. Staff may be prone to thinking that the email they received is safe to leak, since it was sent to a group email address, though the emails are in fact individually marked.

Thus, receiving an email sent to what appears to be a group email address is not a guarantee that an email hasn’t been individually watermarked.

Good News, Bad News

Spatial watermarking can be neutralized through manual transcription. Instead of printing or copy-and-pasting a document, a leaker or the publisher of a leaked document can retype a document into plain-text format; this would get rid of spatial watermarks, as well as other techniques such as font-based watermarking, which would entail sending every recipient an email in a slightly different font, or homoglyph watermarking, which replaces certain characters with lookalike characters.

That’s the good news. However, in addition to “open space” watermarks, text can be watermarked via minute syntactic (structural) as well as semantic (word choice) alterations.

For instance, an example of syntactic watermarking arose when the website Genius, which posts and allows users to annotate song lyrics, suspected that Google was taking lyrics from their site and reproducing them in full in its search results (Google was trying to keep users from clicking away to other sites). Genius watermarked their lyrics with variations of straight and curly single-quote characters, which, when translated to Morse code, spelled out the word “red-handed.” When a search for song lyrics on Google turned up the same pattern of punctuation marks, Google was indeed caught red-handed.

It’s also been reported that Musk has used semantic watermarking techniques. As described by Gawker in 2009, “Musk set out to entrap potential leakers by sending each employee a slightly altered version of an email which he expected would get sent to the media.” Each copy of the email used unique word arrangements; for instance some stated “I am,” while others said “I’m.” The watermarking scheme was foiled when Tesla’s general counsel apparently forwarded to everyone in the company his copy of the email, which meant that staff could now compare the version they had received to the lawyer’s version. They could also just leak the lawyer’s version of the email.

Each copy of the email used unique word arrangements; for instance some stated “I am,” while others said “I’m.”

This case highlights how semantic watermarking would survive manual transcription, though it can be foiled by comparing multiple copies of a given text. However, if it’s not possible to review multiple copies, and there’s a possibility a document has been semantically watermarked, then it’s best to not reproduce the original document in a story, as well as, ideally, not quoting from it, lest the quotation contains part of the semantic watermark.

However, the deployment of a so-called canary trap or barium meal test extends beyond spacing or word alterations in documents. Other tactics involve each person at an organization being presented with a unique document or unique piece of information in a document (say, a supposedly new Twitter feature mentioned only to a suspected leaker). In these cases, a story’s reference to a particular document, or a particular item in the document, could identify the leaker. This highlights the crucial importance of obtaining multiple sources to confirm new information in a leaked document – and this may be tricky as revealing the new piece of information to a second source may compromise the original source if the second one mentions it to someone else at the organization.

Ultimately, a variety of strategies can be used to attempt to safeguard a source from watermarking schemes used by Musk or others, ranging from confirming that the same copy of a document was presented to multiple staff, to transcribing the document, to not quoting from the document, to altogether not mentioning a given document. Though organizations may have a variety of tricks up their sleeve, leakers are far from powerless in this dynamic and have a number of techniques at their disposal to foil watermarking measures.

Nonetheless, it’s important to remember that even the best attempts at foiling watermarks are not foolproof guarantees against source identification, as there are other methods of workplace surveillance, including audits of who accessed leaked documents and video footage of employees copying documents. With Musk recently making threats against would-be leakers, it’s now more important than ever to stay vigilant — even if you don’t work at Twitter.

The post How Elon Musk Says He Catches Leakers at His Companies appeared first on The Intercept.

This post was originally published on The Intercept.

In a change to its anti-doxxing policy made Wednesday, Twitter barred users from sharing a person’s “live” location, a broad, vague, and immediately confusing prohibition. The policy was amended on the same day Twitter banned @ElonJet, an account that tracked owner Elon Musk’s personal private jet, along with the account of its creator, college sophomore Jack Sweeney. Later, the @ElonJet account, but not Sweeney’s private account, was reinstated.

Twitter’s newly revised “Private information and media policy” now forbids users from sharing “live location information, including information shared on Twitter directly or links to 3rd-party URL(s) of travel routes, actual physical location, or other identifying information that would reveal a person’s location, regardless if this information is publicly available.”

The new rule, which an Internet Archive snapshot of the page shows was not present the day before Sweeney and @ElonJet were banned, is at odds with Musk’s gesturing toward free speech absolutism. He claimed that his purchase of the social media giant augured a radically more permissive era for its users — specifically mentioning Sweeney’s account.

On November 6, Musk pledged that he would not ban @ElonJet. “My commitment to free speech extends even to not banning the account following my plane, even though that is a direct personal safety risk,” Musk tweeted. On Wednesday, less than a month later, Musk reversed course entirely: “Real-time posting of someone else’s location violates doxxing policy, but delayed posting of locations are ok.” Hours later, @ElonJet was suddenly back, without explanation.

@ElonJet uses freely available public flight data to chart trips using Musk’s jet, whether he was aboard or not. Virtually every single aircraft in the sky broadcasts such location data through a legally mandated radio transponder. Other flight-tracking accounts created by Sweeney, such as one that tracks the planes of Russian oligarchs, remain offline.

The @ElonJet account had previously attracted Musk’s ire, particularly after Sweeney rejected a $5,000 offer from the world’s then-richest man to voluntarily shutter the account in January.

Late Wednesday afternoon, a Twitter Safety account clarified that tweeting someone’s precise location would be allowed so long as it was “not same-day” — a crucial term left undefined. The account added: “Content that shares location information related to a public engagement or event, such as a concert or political event, is also permitted” — though it’s similarly unclear what exactly fits the definition of a “public engagement or event,” or how the rule could affect news-gathering or the vast volume of ordinary inoffensive speech that merely observes that a given person is currently at a given place.

The total ambiguity of the rule — would it prohibit tweeting a picture you just took of Times Square, thereby disclosing the exact location of every stranger in it? — will give Musk a great deal of latitude in how and when it’s enforced.

The revised policy further says, “If your account is dedicated to sharing someone’s live location, your account will be automatically suspended” — a brand-new rule under which @ElonJet was unceremoniously banned, before being inexplicably later reinstated.

A Twitter spokesperson could not be reached for comment; the company no longer has a communications team.

The post Tweaked Twitter Privacy Rules Would Ban Elon Musk’s Bêtes Noires — or Not appeared first on The Intercept.

This post was originally published on The Intercept.

INTRODUCTION

On 24 June 2022, the United States Supreme Court declared in Dobbs v. Jackson Women’s Health Organization that there is not any constitutional right to abortion, reversing Roe v. Wade and Planned Parenthood of Southeastern Pennsylvania v. Casey—two landmark decisions which had affirmed American’s right to access abortions.

Since the Dobbs decision, there has been much discussion over the widespread use of period tracking apps, which according to a report by Consumer Reports are used by around 50 million women worldwide. It has been cautioned by some advocates that law enforcement in states where abortion is criminalised can and would subpoena the data obtained by these apps to be used as evidence of a terminated pregnancy in an effort to combat self-managed abortions, despite the fact that such abortions are medically identical to miscarriages. Period trackers could, theoretically, share user information with law enforcement if faced with requests.

ABORTION AND PRIVACY RIGHTS IN THE DIGITAL WORLD

It is worth mentioning that the right to privacy and the right to have an abortion are two rights that are intertwined in the legal system of the United States, despite the fact that they may seem to be unrelated at first look. This is because the decision in Roe v. Wade, which was handed down in 1973, states that the right to abortion originates from the constitutional right to privacy that is provided by the due process clause of the Fourteenth Amendment.

It is well-established that the criminalisation of abortion forces women to turn to illegal and frequently dangerous procedures. But in the digital era, restricted access to reproductive care has also bred worries about privacy. Privacy advocates worry that state and local governments may use sensitive information, such as data about menstrual cycles and reproductive health from period-tracking apps, search histories and even text messages to identify and punish women who seek abortions in the wake of the Dobbs ruling. Many of these period tracking applications also work in the AdTech sector and employ business strategies that generate income by selling users’ sensitive information.

The increased criminalisation of reproductive health care has serious ramifications for the lack of robust digital privacy protections for health apps, such as period trackers. In 2019, Privacy International published a study highlighting  how certain popular period tracking applications were sharing the data they acquired with Facebook for advertisement purposes. In a different study, Privacy International requested data from five period tracking applications in order to examine these apps’ adherence to GDPR standards, including the right of access to one’s data and to observe how users’ data was being used and shared. When the researchers tried to access their own data, only two apps responded, and even then, their data was occasionally shared with outside parties. In 2021, a complaint was filed against the widely-used app Flo, alleging that “app events,” or app data shared to third parties for various reasons, had been used to disclose sensitive health information, such as the confirmation of a user’s pregnancy, to third parties. After Dobbs, privacy advocates and women’s rights organisations worry that state officials will now have access to and be free to exploit this data.

These concerns are not unfounded. In one widely reported instance, a Mississippi mother who gave birth to a stillborn child at home was accused of murder because she had looked up abortion medications online. In another instance, prosecutors used a woman’s text messages as evidence that her miscarriage was in fact a self-induced abortion, and she was given a 20-year prison term for feticide. Her conviction was ultimately overturned, but only after she had served three years in prison.

According to a recent investigation by Reveal and The Markup, anti-abortion groups used Facebook’s advertising tools, which collect data from large portions of the internet—including some hospitals—to monitor people seeking abortion services, in spite of Meta’s rules prohibiting the collection of such information. According to the investigation, data gathered by the organisations was also shared with various anti-abortion marketing businesses, allowing them to target advertisements to “abortion-minded women.” While there have been no confirmed cases thus far of period trackers being used in prosecutions of abortions, the data collected by digital health applications, such as period trackers, is potentially subject to subpoena and therefore could be used by law enforcement as evidence of abortion in jurisdictions where the practise is now illegal.

SCOPE OF EXISTING DATA PRIVACY LAWS

Before the Dobbs decision was finalised, lawmakers urged Google and the Federal Trade Commission to guarantee that in the event of such a decision, data for online patients seeking care would be protected. The letters were sent after Politico reported on a draft of the Dobbs ruling that had been leaked, indicating that the Supreme Court was preparing to overturn Roe. In many areas of the United States, a routine, safe medical procedure was swiftly made illegal by the repeal of federal abortion rights, and routine medical information was transformed into evidence that could be used against people who might have had an abortion.

Medical privacy laws in the United states, the most significant of which is the Health Insurance Portability and Accountability Act (HIPAA), do not prevent subpoenas or warrants for medical records data, and as a result, doctors are permitted to reveal medical information if they have reason to believe that a crime has been committed. The majority of patients do not possess their own medical records, and patients have clear ownership rights over their medical records in only one state, New Hampshire. In  other states, the laws expressly state that hospitals or medical providers are the owners of the records, though people must be able to view their own complete medical records in accordance with HIPAA. One could believe that the HIPAA safeguards the medical information of abortion-seeking women. However, the law is deficient in this area, and patients’ privacy is not completely protected.

Experts in health law, Kayte Spector-Bagdady and Michelle Mello, suggested three scenarios where HIPAA would fall short of protecting patients’ privacy. These include using medical records of a patient to prove that they are seeking an abortion, using medical facility records to prove that the medical facility is at fault and using online activity data to prove that a patient is seeking an abortion. US senators recently urged the Department of Human and Health Services (HHS) to address this problem and safeguard patients’ reproductive healthcare information. After that, the HHS Office of Civil Rights published guidance outlining ways that people could protect their health information while using cell phones and tablets. The guidance also advised healthcare professionals that the privacy rule under HIPAA would prevent them from disclosing information about abortion-related incidents unless required to do so by State law. Additionally, it was stated in the guidance that healthcare providers were not required to divulge a patient’s personal health information absent a court order or subpoena from law enforcement. The HHS has been urged by the US senators to take additional action, including using their administrative authority to update the HIPAA privacy rule, identify the covered entities, restrict the circumstances under which these entities may share information about reproductive health or abortion, and make it clear that healthcare information cannot be shared with law enforcement agencies that may be targeting women seeking abortions.

Surprisingly, because they do not qualify as covered companies, period-tracking apps are not subjected to HIPAA regulations, which enables them to freely monetise the data they gather. As a result, many menstrual apps in the US have been found to exchange data with outside parties, and advertisers use information like a user’s desire for a child or desire to avoid having one to display advertisements for fertility or abortion clinics. Thus, the lack of adequate regulations concerning the use of health data by period tracking apps raises serious concerns about users’ digital privacy in the wake of the Dobbs decision.

CONCLUSION

The United States Supreme Court’s decision in Dobbs v. Jackson Women’s Health Organization highlights the need for  tech corporations to recognise their own power and the elevated significance of user privacy. We currently live in a world where we leave unparalleled digital fingerprints. The data-sharing tactics of period-tracking apps can exacerbate issues for women and those who procreate, but they can also aid in the diagnosis of diseases, like polycystic ovarian syndrome. Some apps assist in the creation of reports that can be shared with medical practitioners, which makes opting out more difficult and unpleasant, especially for users without access to health insurance or other forms of medical assistance. Today, Protected Health Information (PHI) is gathered, stored, communicated and guarded in quite different ways than it was  when HIPAA was passed more than 25 years ago. As new technological advancements and data storage techniques take hold, providers and patients alike are discovering that HIPAA’s patient privacy protections are less comprehensive than previously believed.

Following are  suggestions in order to address some of the problems with period tracking software collecting personal data:

• Strengthen the security system: Companies are required to inform the period tracking application users of the security precautions and procedures that will be used to protect their data, which might include conducting compliance reports of the data laws prevailing in the countries where their services are available.
• Users must also be made aware of any breaches or hacks and provided with information about how much their data has been impacted, which might also include sending data breach notifications to its users.
• User rights should not be restricted, and data fiduciaries should not have the authority to refuse users’ requests outright.

 

REFERENCE LIST

1. Michela Moscufo, Parks, M., Jeca Taudte and DiMartino, J. (2022). Period-tracking apps may help prosecute users, advocates fear. [online] ABC News. Available at: https://abcnews.go.com/Health/abortion-advocates-fear-period-tracking-apps-prosecute-abortion/story?id=85925714 [Accessed 28 Nov. 2022].
2. ‌Rosato, D. (2020). What Your Period Tracker App Knows About You. [online] Consumer Reports. Available at: https://www.consumerreports.org/health-privacy/what-your-period-tracker-app-knows-about-you-a8701683935/ [Accessed 28 Nov. 2022].
3. ‌Garamvolgyi, F. (2022). Why US women are deleting their period tracking apps. [online] the Guardian. Available at: https://www.theguardian.com/world/2022/jun/28/why-us-woman-are-deleting-their-period-tracking-apps [Accessed 28 Nov. 2022].
4. ‌Justia Law. (2022). Planned Parenthood of Southeastern Pa. v. Casey, 505 U.S. 833 (1992). [online] Available at: https://supreme.justia.com/cases/federal/us/505/833/ [Accessed 28 Nov. 2022].
5. ‌Justia Law. (2022). Roe v. Wade, 410 U.S. 113 (1973). [online] Available at: https://supreme.justia.com/cases/federal/us/410/113/ [Accessed 28 Nov. 2022].
6. ‌SUPREME COURT OF THE UNITED STATES (2021). Dobbs v. Jackson Women’s Health Organization. [online] Available at: https://www.supremecourt.gov/opinions/21pdf/19-1392_6j37.pdf.
7. ‌Morse, J. (2022). What Is Roe v. Wade ? And Why Experts Think Your Privacy Is At Risk Should It Fall. [online] Mashable India. Available at: https://in.mashable.com/tech/31277/what-is-roe-v-wade-and-why-experts-think-your-privacy-is-at-risk-should-it-fall [Accessed 28 Nov. 2022].
8. ‌Amnesty International. (2022). Abortion is a human right. [online] Available at: https://www.amnesty.org/en/what-we-do/sexual-and-reproductive-rights/abortion-facts/ [Accessed 28 Nov. 2022].
9. ‌AP (2022). In a post Roe world Big Tech policies regarding abortion could impact user privacy. [online] Thehindu.com. Available at: https://www.thehindu.com/sci-tech/technology/roe-vs-wade-america-abortion-pregnancy-big-tech-meta-facebook/article65756820.ece [Accessed 28 Nov. 2022].
10. ‌Alaattinoğlu, D. (2022). Rethinking Explicit Consent and Intimate Data: The Case of Menstruapps. Feminist Legal Studies, [online] 30(2), pp.157–179. doi:10.1007/s10691-021-09486-y.
11. ‌Privacy International. (2020). No Body’s Business But Mine: How Menstruation Apps Are Sharing Your Data. [online] Available at: https://privacyinternational.org/long-read/3196/no-bodys-business-mine-how-menstruations-apps-are-sharing-your-data [Accessed 28 Nov. 2022].
12. ‌Privacy International. (2020). We asked five menstruation apps for our data and here is what we found… [online] Available at: https://privacyinternational.org/long-read/4316/we-asked-five-menstruation-apps-our-data-and-here-what-we-found [Accessed 28 Nov. 2022].
13. ‌GDPR.eu. (2019). Everything you need to know about GDPR compliance – GDPR.eu. [online] Available at: https://gdpr.eu/compliance/ [Accessed 28 Nov. 2022].
14. ‌Federal Trade Commission. (2021). Health app broke its privacy promises by disclosing intimate details about users. [online] Available at: https://www.ftc.gov/business-guidance/blog/2021/01/health-app-broke-its-privacy-promises-disclosing-intimate-details-about-users [Accessed 28 Nov. 2022].
15. ‌Malek, L.A., Jain, P. and Song, K. (2022). Pandora’s Box of Data Privacy at Risk With Abortion Ruling. [online] @BLaw. Available at: https://news.bloomberglaw.com/health-law-and-business/pandoras-box-of-data-privacy-at-risk-with-abortion-ruling [Accessed 28 Nov. 2022].
16. ‌Gal, U. (2022). Post Roe, women in America are right to be concerned about digital surveillance – and it’s not just period-tracking apps. [online] The Conversation. Available at: https://theconversation.com/post-roe-women-in-america-are-right-to-be-concerned-about-digital-surveillance-and-its-not-just-period-tracking-apps-185865 [Accessed 28 Nov. 2022].
17. ‌Zakrzewski, C., Verma, P. and Parker, C. (2022). Texts, web searches about abortion have been used to prosecute women. [online] Washington Post. Available at: https://www.washingtonpost.com/technology/2022/07/03/abortion-data-privacy-prosecution/ [Accessed 28 Nov. 2022].
18. ‌Christophe Haubursin (2015). An Indiana woman is facing 20 years in prison for ‘feticide’. [online] Vox. Available at: https://www.vox.com/2015/4/3/8336863/an-indiana-woman-is-facing-20-years-in-prison-for-feticide [Accessed 28 Nov. 2022].
19. ‌Oldham, G. and Mehrotra, D. (2022). Facebook Enables Anti-Abortion Clinics to Collect Data on Would-Be Patients. [online] Truthout. Available at: https://truthout.org/articles/facebook-enables-anti-abortion-clinics-to-collect-data-on-would-be-patients/ [Accessed 28 Nov. 2022].
20. ‌Suciu, P. (2022). Facebook Has Been Targeting Users Who Visited Pregnancy Center Sites – How Could That Data Be Used? Forbes. [online] 11 Jul. Available at: https://www.forbes.com/sites/petersuciu/2022/07/10/facebook-has-been-targeting-users-who-visited-pregnancy-center-sites–how-could-that-data-be-used/?sh=104bf3727c53 [Accessed 28 Nov. 2022].
21. ‌Abrams, A. and Bergengruen, V. (2022). Anti-Abortion Pregnancy Centers Are Collecting Troves of Data That Could Be Weaponized Against Women. [online] Time. Available at: https://time.com/6189528/anti-abortion-pregnancy-centers-collect-data-investigation/ [Accessed 28 Nov. 2022].
22. ‌Feiner, L. (2022). Democrats urge Google to stop collecting location data that could be used to identify people seeking abortions. [online] CNBC. Available at: https://www.cnbc.com/2022/05/24/google-data-collection-could-endanger-abortion-seekers-say-dems.html [Accessed 28 Nov. 2022].
23. ‌Feiner, L. (2022). Democratic senators concerned about phone location data being used to track people seeking abortions. [online] CNBC. Available at: https://www.cnbc.com/2022/05/20/democrats-urge-ftc-protect-data-privacy-for-people-seeking-abortions.html [Accessed 28 Nov. 2022].
24. ‌POLITICO. (2022). Exclusive: Supreme Court has voted to overturn abortion rights, draft opinion shows. [online] Available at: https://www.politico.com/news/2022/05/02/supreme-court-abortion-draft-opinion-00029473 [Accessed 28 Nov. 2022].
25. ‌Brandom, R., Wetsman, N., Corin Faife and Mary Beth Griggs (2022). The biggest privacy risks in seeking abortion care in post-Roe America. [online] The Verge. Available at: https://www.theverge.com/23185081/abortion-data-privacy-roe-v-wade-dobbs-surveillance-period-tracking [Accessed 28 Nov. 2022].
26. ‌HEALTH INSURANCE PORTABILITY AND ACCOUNTABILITY ACT OF 1996. (1996). [online] Available at: https://www.govinfo.gov/content/pkg/PLAW-104publ191/pdf/PLAW-104publ191.pdf.
27. ‌Healthinfolaw.org. (2012). Who Owns Medical Records: 50 State Comparison | Health Information & the Law. [online] Available at: http://www.healthinfolaw.org/comparative-analysis/who-owns-medical-records-50-state-comparison [Accessed 28 Nov. 2022].
28. ‌Spector-Bagdady, K. (2021). Governing secondary research use of health data and specimens: the inequitable distribution of regulatory burden between federally funded and industry research. Journal of Law and the Biosciences, [online] 8(1). doi:10.1093/jlb/lsab008.
29. ‌HHS.gov. (2022). U.S. Department of Health & Human Services (HHS). [online] Available at: https://www.hhs.gov/ [Accessed 28 Nov. 2022].
30. ‌Office (2022). HIPAA Privacy Rule and Disclosures of Information Relating to Reproductive Health Care. [online] HHS.gov. Available at: https://www.hhs.gov/hipaa/for-professionals/privacy/guidance/phi-reproductive-health/index.html [Accessed 28 Nov. 2022].

This post was originally published on LSE Human Rights.

by James Bandler

ProPublica is a nonprofit newsroom that investigates abuses of power. Sign up to receive our biggest stories as soon as they’re published.

In early February 2016, the security gate at a U.S. military base near Washington, D.C., swung open to admit a Navy doctor accompanying a pair of surprising visitors: two artificial intelligence scientists from Google.

In a cavernous, temperature-controlled warehouse at the Joint Pathology Center, they stood amid stacks holding the crown jewels of the center’s collection: tens of millions of pathology slides containing slivers of skin, tumor biopsies and slices of organs from armed service members and veterans.

Standing with their Navy sponsor behind them, the Google scientists posed for a photograph, beaming.

Mostly unknown to the public, the trove and the staff who study it have long been regarded in pathology circles as vital national resources: Scientists used a dead soldier’s specimen that was archived here to perform the first genetic sequencing of the 1918 Flu.

Google had a confidential plan to turn the collection of slides into an immense archive that — with the help of the company’s burgeoning, and potentially profitable, AI business — could help create tools to aid the diagnosis and treatment of cancer and other diseases. And it would seek first, exclusive dibs to do so.

“The chief concern,” Google’s liaison in the military warned the leaders of the repository, “is keeping this out of the press.”

More than six years later, Google is still laboring to turn this vast collection of human specimens into digital gold.

At least a dozen Defense Department staff members have raised ethical or legal concerns about Google’s quest for service members’ medical data and about the behavior of its military supporters, records reviewed by ProPublica show. Underlying their complaints are concerns about privacy, favoritism and the private use of a sensitive government resource in a time when AI in health care shows both great promise and risk. And some of them worried that Google was upending the center’s own pilot project to digitize its collection for future AI use.

Pathology experts familiar with the collection say the center’s leaders have good reason to be cautious about partnerships with AI companies. “Well designed, correctly validated and ethically implemented [health algorithms] could be game-changing things,” said Dr. Monica E de Baca, chair of the College of American Pathologists’ Council on Informatics and Pathology Innovation. “But until we figure out how to do that well, I’m worried that –knowingly or unknowingly –there will be an awful lot of snake oil sold.”

When it wasn’t chosen to take part in JPC’s pilot project, Google pulled levers in the upper reaches of the Pentagon and in Congress. This year, after lobbying by Google, staff on the House Armed Services Committee quietly inserted language into a report accompanying the Defense Authorization Act that raises doubts about the pathology center’s modernization efforts while providing a path for the tech giant to land future AI work with the center.

Pathology experts call the JPC collection a national treasure, unique in its age, size and breadth. The archive holds more than 31 million blocks of human tissue and 55 million slides. More recent specimens are linked with detailed patient information, including pathologist annotations and case histories. And the repository holds many examples of “edge cases” — diseases so vanishingly rare that many pathologists never see them.

Human tissue samples from 1917 and 1918 stored in paraffin are part of the Joint Pathology Center’s collection, which contains more than 31 million tissue blocks and 55 million slides. (Linda Davidson/The Washington Post via Getty Images)

Google sought to gather so many identifying details about the specimens and patients that the repository’s leaders feared it would compromise patients’ anonymity. Discussions became so contentious in 2017 that the leaders of the JPC broke them off.

In an interview with ProPublica, retired Col. Clayton Simon, the former director of the JPC, said Google wanted more than the pathology center felt it could provide. “Ultimately, even through negotiations, we were unable to find a pathway that we legally could do and ethically should do,” Simon said. “And the partnership dissolved.”

But Google didn’t give up. Last year, the center’s current director, Col. Joel Moncur, in response to questions from DOD lawyers, warned that the actions of Google’s chief research partner in the military “could cause a breach of patient privacy and could lead to a scandal that adversely affects the military.”

Joel Moncur (Kate Copeland for ProPublica)

Google has told the military that the JPC collection holds the “raw materials” for the most significant biotechnology breakthroughs of this decade — “on par with the Human Genome Project in its potential for strategic, clinical, and economic impact.”

All of that made the cache an alluring target for any company hoping to develop health care algorithms. Enormous quantities of medical data are needed to design algorithmic models that can identify patterns a pathologist might miss — and Google and other companies are in a race to gather them. Only a handful of tech companies have the scale to scan, store and analyze a collection of this magnitude on their own. Companies that have submitted plans to compete for aspects of the center’s modernization project include Amazon Web Services, Cerner Corp. and a host of small AI companies.

But no company has been as aggressive as Google, whose parent company, Alphabet, has previously drawn fire for its efforts to gather and crunch medical data. In the United Kingdom, regulators reprimanded a hospital in 2017 for providing data on more than 1.6 million patients, without their understanding, to Alphabet’s AI unit, DeepMind. In 2019, The Wall Street Journal reported that Google had a secret deal, dubbed “Project Nightingale,” with a Catholic health care system that gave it access to data on millions of patients in 21 states, also without the knowledge of patients or doctors. Google responded to the Journal story in a blog post that stated that patient data “cannot and will not be combined with any Google consumer data.”

In a statement, Ted Ladd, a Google spokesperson, attributed the ethics complaints associated with its efforts to work with the repository to an “inter-agency issue” and a “personnel dispute.”

“We had hoped to enable the JPC to digitize its data and, with its permission, develop computer models that would enable researchers and clinicians to improve diagnosis for cancers and other illnesses,” Ladd said, noting that all of Google’s health care partnerships involve “the strictest controls” over data. “Our customers own and manage their data, and we cannot — and do not — use it for any purpose other than explicitly agreed upon by the customer,” Ladd said.

In response to questions from ProPublica, the JPC said none of its de-identified data would be shared during its modernization process unless it met the ethical, regulatory, and legal approvals needed to ensure it was done in the right way.

“The highest priority of the JPC’s digital transformation is to ensure that any de-identified digital slides are used ethically and in a manner that protects patient privacy and military security,” the JPC said.

But some fear that even these safeguards might not be enough. Steven French, a DOD cloud computing engineer assigned to the project, said he was dismayed by the relentlessness of Google’s advocates in the department. Lost in all their discussions about the speed, scale and cost-saving benefits associated with working with Google seemed to be concerns for the interests of the service members whose tissue was the subject of all this maneuvering, French told ProPublica.

“It felt really bad to me,” French said. “Like a slow crush towards the inevitability of some big tech company monetizing it.”

The JPC certainly does need help from tech companies. Underfunded by Congress and long neglected by the Pentagon, it is vulnerable to offers from well-funded rescuers. In spite of its leaders’ pleas, funding for a full-scale modernization project has never materialized. The pathology center’s aging warehouses have been afflicted with water leaks and unwelcome intruders: a marauding family of raccoons.

The story of the pathology center’s long, contentious battle with Google has never been told before. ProPublica’s account is based on internal emails, presentations and memos, as well as interviews with current and former DOD officials, some of whom asked not to be identified because they were not authorized to discuss the matter or for fear of retribution.

Google’s Private Tour

In December 2015, Google began its courtship of the JPC with a bold, unsolicited proposal. The messenger was a junior naval officer, Lt. Cmdr. Niels Olson.

“I’m working with Google on a project to apply machine learning to medical imaging,” Olson wrote to the leaders of the repository. “And it seems like we are at the stage where we need to figure exactly what JPC has.”

Niels Olson (Kate Copeland for ProPublica)

A United States Naval Academy physics major and Tulane medical school graduate, Olson worked as a clinical and anatomical pathology resident at the Naval Medical Center in San Diego.

With digitized specimen slides holding massive amounts of data, pathology seemed ripe for the coming AI revolution in medicine, he believed. Olson’s own urgency was heightened in 2014 when his father was diagnosed with prostate cancer.

That year, Olson teamed up with scientists at Google to train software to recognize suspected cancer cells. Google supplied expertise including AI scientists and high-speed, high-resolution scanners. The endeavor had cleared all privacy and review board hurdles. They were scanning Navy patients’ pathology slides at a furious clip, but they needed a larger data set to validate their findings.

Enter the JPC’s archive. Olson learned about the center in medical school. In his email to its leaders in December 2015, Olson attached Google’s eight-page proposal.

Google offered to start the operation by training algorithms with already digitized data in the repository. And it would do this early work “with no exchange of funds.” These types of partnerships free the private parties from having to undergo a competitive bidding process.

Google promised to do the work in a manner that balanced “privacy and ethical considerations.” The government, under the proposal, would own and control the slides and data.

Olson typed a warning: “This is under a non-disclosure agreement with Google, so I need to ask you, do please handle this information appropriately. The chief concern is keeping this out of the press.”

Senior military and civilian staff at the pathology center reacted with alarm. Dr. Francisco Rentas, the head of the archive’s tissue operations, pushed back against the notion of sharing the data with Google.

“As you know, we have the largest pathology repository in the world and a lot of entities will love to get their hands on it, including Google competitors. How do we overcome that?” Rentas asked in an email.

Olson, center, and Google scientists Martin Stumpe and Lily Peng took a private tour of the JPC collection in 2016. (Obtained by ProPublica)

Other leaders had similar reactions. “My concerns are raised when I’m advised to not disclose what seems to be a contractual relationship to the press,” one of the top managers at the pathology center, Col. Edward Stevens, told Olson. Stevens told Olson that giving Google access to this information without a competitive bid could result in litigation from the company’s competitors. Stevens asked: “Does this need to go through an open-source bid?”

But even with these concerns, Simon, the pathology center’s director, was intrigued enough to continue discussions. He invited Olson and Google to inspect the facility.

The warehouse Olson and the Google scientists entered could have served as a set for the final scene of “Raiders of Lost Ark.”

Pathology slides were stacked in aisle canyons, some towering two stories. The slides were arranged in metal trays and cardboard boxes. To access tissue samples, the repository used a retrieval system similar to those found in dry cleaners. The pathology center had just a handful of working scanners. At the pace they were going, it would take centuries to digitize the entire collection.

One person familiar with the repository likened it to the Library of Alexandria, which held the largest archive of knowledge in the ancient world. Myth held that the library was destroyed in a cataclysmic fire lit by Roman invaders, but historians believe the real killer was gradual decay and neglect over centuries.

The JPC’s collection is the largest biorepository on the planet. (Linda Davidson/The Washington Post via Getty Images)

The military’s tissue library had already played an important role in the advancement of medical knowledge. Its birth in 1862 as the Army Medical Museum was grisly. In a blandly written order in the midst of the Civil War, the Army surgeon general instructed surgeons “diligently to collect and preserve” all specimens of “morbid anatomy, surgical or medical, which may be regarded as valuable.”

Soon the museum’s curator was digging through battlefield trenches to find “many a putrid heap” of hands, feet and other body parts ravaged by disease and war. He and other doctors shipped the remains to Washington in whiskey-filled casks.

Over the next 160 years, the tissue collection outgrew several headquarters, including Washington’s Ford Theater and a nuclear-bomb-proof building near the White House. But the main mission — identifying, studying and reducing the calamitous impact of illnesses and injuries afflicting service members — has remained unchanged in times of war and peace. Each time a military or veterans’ hospital pathologist sent a tissue sample to the pathology center for a second opinion, it was filed away in the repository.

As the archive expanded, the repository’s prestige grew. Its scientists spurred advances in microscopy, cancer and tropical disease research. An institute pathologist named Walter Reed proved that mosquitoes transmit yellow fever, an important discovery in the history of medicine.

For much of its modern history, in addition to serving military and veterans hospitals, the center also provided civilian consultations. The work with elite teaching hospitals gave the center a luster that helped it attract and retain top pathologists.

Congress and DOD leaders questioned why the military should fund civilian work that could be done elsewhere. In 2005, under the congressionally mandated base closure act, the Pentagon ordered the organization running the repository to shut down. The organization reopened with a different overseer, tasked with a narrower, military-focused mission. Uncertainty about the organization’s future caused many top pathologists to leave.

In its first pitch to the repository’s leaders, Google pointedly mentioned a book-length Institute of Medicine report on the repository that stated that “wide access” to the archive’s materials would promote the “public good.” The biorepository wasn’t living up to its potential, Google said, noting that “no major efforts have been underway to fix the problem.”

Following the tour, a Google scientist prepared a list of clinical, demographic and patient information it sought from the repository. The list included “must haves” — case diagnoses; pathology and radiology images; information on gender and ethnicity; and birth and death dates — as well as “high-value” patient information, including comorbidities, subsequent hospitalizations and cause of death.

This troubled the JPC’s director. “We felt very, very concerned about giving too much data to them,” Simon told ProPublica, “because too much data could identify the patient.”

There were other aspects about Google’s offer that made it “very unfavorable to the federal government,” Simon later told his successor, according to an email reviewed by ProPublica.

In exchange for scanning and digitizing the slide collection at its own expense, Google sought “exclusive access” to the data for at least four years.

The other deal-breaker was Google’s requirement that it be able to charge the government to store and access the digitized information, a huge financial commitment. Simon did not have the authority to commit the government to future payments to a company without authorization from Congress.

Today, Ladd, the Google spokesperson, disputes the claim that its proposal would have been unfavorable to the government. “Our goal was to help the government digitize the data before it physically deteriorates.”

Ladd said Google sought exclusive access to the data during the early stages of the project, so that it could scan the de-identified samples and perform quality-control measures on the data prior to handing it back to the JPC.

Niels Olson, who spearheaded the project for the Navy in 2016, declined requests for interviews with ProPublica. But Jackson Stephens, a friend and lawyer who is representing Olson, said Olson had always followed the Institutional Review Board process and worked to anonymize patient medical data before it was used in research or shared with a third party.

“Niels takes his oath to the Constitution and his Hippocratic oath very seriously,” Stephens said. “He loves science, but his first duty of care is to his patients.”

Google’s relentlessness in 2017, too, spooked the repository’s leaders, according to an email reviewed by ProPublica. Google’s lawyer put “pressure” on the head of tissue operations to sign the agreement, which he declined to do. Leaders of the center became “uncomfortable” and discontinued discussions, according to the DOD email.

Though he banged on doors in the Pentagon and Congress, Simon was not able to convince the Obama administration to include the JPC in then-Vice President Joe Biden’s Cancer Moonshot. Simon left the JPC in 2018, his hopes for a modernization of the library dashed. But then a Pentagon advisory board got wind of the JPC collection, and everything changed.

“The Smartest People on Earth”

In March of 2020, the Defense Innovation Board announced a series of recommendations to digitize the JPC collection. The board called for a pilot project to scan a large initial batch of slides — at least 1 million in the first year — as a prelude to the massive undertaking of digitizing all 55 million slides.

“My worldview was that this should be one of the highest priorities of the Defense Department,” William Bushman, then acting deputy undersecretary of personnel and readiness, told ProPublica. “It has the potential to save more lives than anything else being done in the department.”

As the pathology center prepared to launch its pilot, the staff talked about a scandal that occurred just 40 miles north.

Henrietta Lacks was a Black woman who died of cancer in 1951 while being treated at Baltimore’s Johns Hopkins Hospital. Without her or her family’s knowledge or consent, and without compensation, her cells were replicated and commercialized, leading to groundbreaking advances in medicine but also federal reforms on the use of patient cells for research.

A photo of Henrietta Lacks sits in the living room of her grandson, Ron Lacks. (Jonathan Newton/The Washington Post via Getty Images)

Like Lacks’ cancer cells, every specimen in the archive, the JPC team knew, represented its own story of human mortality and vulnerability. The tissue came from veterans and current service members willing to put their lives on the line for their country. Most of the samples came from patients whose doctors discovered ominous signs from biopsies and then sent the specimens to the center for second opinions. Few signed consent forms agreeing to have their samples used in medical research.

The pathology center hired two experts in AI ethics to develop ethical, legal and regulatory guidelines. Meanwhile, the pressure to cooperate with Google hadn’t gone away.

In the summer of 2020, as COVID-19 surged across the country, Olson was stationed at a naval lab in Guam, working on an AI project to detect the coronavirus. That project was managed by a military group based out of Silicon Valley known as the Defense Innovation Unit, a separate effort to speed the military’s development and adoption of cutting-edge technology. Though the group worked with many tech companies, it had gained a reputation for being cozy with Google. The DIU’s headquarters in Mountain View, California, sat just across the street from the Googleplex, the tech giant’s headquarters. Olson joined the group officially that August.

Olson’s COVID-19 work earned him Navy Times’ coveted Sailor of the Year award as well as the attention of a man who would become a powerful ally in the DOD, Thomas “Pat” Flanders.

Flanders was the chief information officer of the sprawling Defense Health Agency, which oversaw the military’s medical services, including hospitals and clinics. A garrulous Army veteran, Flanders questioned the wisdom of running the pilot project without first getting funding to scan all of the 55 million slides. He wanted the pathology staff to hear about the work Olson and Google had done scanning pathology slides in San Diego and see if a similar public-private partnership could be forged with the JPC.

Over the objections of Moncur, the JPC’s director, Flanders insisted on having Olson attend all the pathology center’s meetings to discuss the pilot, according to internal emails.

In August 2020, the JPC published a request for information from vendors interested in taking part in the pilot project. The terms of that request specified that no feedback would be given to companies about their submissions and that telephone inquiries would not be accepted or acknowledged. Such conversations could be seen as favoritism and could lead to a protest by competitors who did not get this privilege.

But Flanders insisted that meeting Google was appropriate, according to Moncur’s statements to DOD lawyers.

In a video conference call, Flanders told the Google representatives they were “the smartest people on earth” and said he couldn’t believe he was “getting to meet them for free,” according to written accounts of the meeting provided to DOD lawyers.

Flanders asked Google to explain its business model, saying he wanted to see how both the government and company might profit from the center’s data so that he could influence the requirements on the government side — a remark that left even the Google representatives “speechless,” according to a compilation of concerns raised by DOD staffers.

To Moncur and others in attendance, Flanders was actively negotiating with Google, according to Moncur’s statement to DOD lawyers.

To the astonishment of the center staff, Flanders asked for a second meeting between Google and the JPC team.

Concern about Flanders’ conduct echoed in other parts of the DOD. A lawyer for Defense Digital Service, a team of software engineers, data scientists and product managers assigned to assist on the project, wrote that Flanders ignored legal warnings. He described Flanders as a “cowboy” who in spite of warnings about his behavior was not likely “to fall out of love with Google.”

In an interview with ProPublica, Flanders disputed claims that he was biased toward Google. Flanders said his focus has always been on scanning and storing the slides as quickly and economically as possible. As for his lavish praise of Google, Flanders said he was merely trying to be “kind” to the company’s representatives.

“People took offense to that,” Flanders said. “It’s just really pettiness on the part of people who couldn’t get along, honestly.”

A spokesperson for the Defense Health Agency said it was “totally appropriate” for Flanders to ask Google about its business model. “This is part of market research,” the spokesperson wrote, adding that no negotiation occurred at the meeting and that all government stakeholders had been invited to attend.

Moncur referred calls to a JPC spokesperson. A spokesperson for the JPC said in a statement that “Moncur was concerned about meeting with vendors during the RFI period.”

“An Arm of Google”

In late 2020, the modernization team received more troubling news. In a slide presentation for the JPC describing other AI work with Google and the military, Olson disclosed that the company had “made offers of employment, which I have declined.” But then he suggested the offer might be revived in the future, writing, “we mutually agreed to table the matter.” He said he had “no other conflicts of interest to declare.” Google told ProPublica it had never directly made Olson a job offer, though a temp agency it used did.

More facts surfaced. Olson also had a Google corporate email address. And he had access to Google corporate files, according to internal communications from concerned DOD staff members. Google said it is common for its research partners in the government to have these privileges.

“I am more worried than ever that DIU’s influence will destroy this acquisition,” a DOD lawyer wrote, referring to efforts to find vendors for the pilot project. He called DIU “essentially an arm of Google.”

At the time, a DIU lawyer defended Olson. The lawyer said Olson had “no further conflict of interest issues” and had done nothing improper because the job offer had been made three years earlier, in 2017. An ethics officer at the DOD Standards of Conduct Office agreed.

Today, a spokesperson in the Office of the Secretary of Defense told ProPublica the department was committed to modernizing the repository “while carefully observing all applicable legal and ethical rules.”

Olson’s friend and lawyer, Stephens, said Olson had been upfront, disclosing the job offer to the innovation unit’s lawyer as well as in the conflict-of-interest section of his slide presentation. He said Olson had declined the offer, which was withdrawn. “He’s not some kind of Google secret agent.”

Stephens said the JPC would have been much further down the road had it cooperated with Olson. Stephens said it became apparent to Olson that Moncur was “essentially ignoring” a “gold mine that could help a lot of people.”

“Niels is the tenacious doctor who is just trying to do the science and build a coalition of partners to get this thing done,” Stephens said. “I think he’s the hero of this story.”

Google Turns to Congress

In 2021, the pathology center selected one of the most prestigious medical institutions in the world, Johns Hopkins — which plans to erect a building honoring Henrietta Lacks — to assist it in scanning slides. It picked two small technology companies to start building tools to let pathologists search the archive.

Google wanted to be selected, and in a confidential proposal, it offered to help the repository build up its own slide-scanning capabilities.

When Google was not selected for the pilot project, the company went above the JPC leaders’ heads. Google claimed in a letter to Pentagon leaders that the company had been unfairly excluded from “full and open competition.” In that August 2021 letter, Google argued that the nation’s security was at stake. It asked the DOD to “consider allowing Google Cloud” and other providers to compete to ensure the “nation’s ability to compete with China in biotechnology.”

Time was of the essence, Google warned. “The physical slides at the JPC are degrading rapidly each day. … Without further action, the slides will continue to degrade and some may ultimately be damaged beyond repair.”

Google stepped up its advocacy campaign. The company deployed a lobbying firm, the Roosevelt Group — which boasts of its ability to “leverage” its connections to secure federal business opportunities to its clients — to raise doubts about the JPC’s pilot project. Their efforts worked. In little-noticed language in a report written to accompany the 2023 Defense Authorization Act, the House Armed Services Committee expressed its concern about the speed of the scanning process and the choice of technology, which the committee claimed would not allow the “swift digitization of these deteriorating slides.”

The committee had its own ideas of how the pathology center’s work should be carried out, suggesting that the center work in tandem with the DIU, using an augmented reality microscope whose software was engineered by Google.

In a statement, the Roosevelt Group told ProPublica it was “proud” of its work for Google. The firm said it helped the company “educate professional staff of the House and Senate Armed Services Committees over concerns about the lack of an open procurement process for digitization of slides.” The group chided DOD officials for being “unwilling to provide answers to Congress around the lack of progress on the JPC digitization effort.”

The pathology center staff was dismayed by the committee’s recommendations that it work with Olson’s group.

In a video conference meeting late last summer with Armed Services Committee staff, the leaders of the pathology center attempted to rebut the House committee report. The JPC’s work was going as planned, they said, noting that a million slides had been scanned. And the pathology center was collaborating with the National Institutes of Health to develop AI tools to help predict prognoses for cancer treatments.

The House Armed Services Committee ordered Pentagon leaders to “conduct a comprehensive assessment” on the digitization effort and to provide a briefing to the committee on its findings by April 1, 2023.

In a statement in response to ProPublica’s questions about the bill, Ladd, the Google spokesperson, acknowledged the company’s influence efforts on Capitol Hill. “We frequently provide information to congressional staff on issues of national importance,” Ladd said. The statement confirmed that the company suggested “language be inserted” into the 2023 Defense Authorization Act calling for a “comprehensive assessment” of the digitization effort.

“Despite efforts from Google and many at the Department of Defense, our work with JPC unfortunately never got off the ground, and the physical repository of pathology slides continues to deteriorate,” Ladd said. “We remain optimistic that if the repository could be properly digitized, it would save many American lives, including those of our service members.”

On this last point, even Google’s critics are in accord. A properly funded project would cost taxpayers a few hundred million dollars — a minuscule portion of the $858 billion defense budget and a small price if the lifesaving potential of the collection is realized.

Last year, as tensions grew with Google, the modernization team at the repository launched a publicity campaign to call attention to the project and the high ethical stakes.

An entire panel discussion was devoted to the JPC effort at the 2021 South by Southwest conference. “This is a once in a lifetime opportunity, and I want to make sure we do it right, we do it responsibly and we do it ethically,” said Steven French, the DOD cloud computing engineer assigned to assist the repository.

Then without mentioning Google’s name, he added a Shakespearean barb. “There’s plenty of vendors, plenty of companies, plenty of people,” French said, “who are more than willing to do this and extract a pound of flesh from us in the process.”

Additional image credits: Duncan1890, Cultura RM Exclusive/PhotoStock-Israel, Rob Jones III, Kampee Patisena, Steve Gschmeissner/Science Photo Library, Sebastian Condrea, Jason Edwards, undefined undefined, Mikroman6, Trifonov_Evgeniy, Zoranm, Wladimir Bulgar/Science Photo Library, Michael Burrell, DanielBendjy, John Parrot/Stocktrek Images, PansLaos, SDI Productions, George Marks, Carlofranco, Tetra Images, Leonello Calvetti/Science Photo Library, Mashuk, and Thepalmer/Getty Images

Doris Burke contributed research.

This post was originally published on Articles and Investigations – ProPublica.