Daily Archives: February 14, 2024

New evidence changes key ideas about Earth’s climate history – it wasn’t that hot

Posted on by

A new study published in Science resolves a long-standing scientific debate, and it stands to completely change the way we think about Earth’s climate evolution.

The research debunks the idea that Earth’s surface (across land and sea) has experienced really hot temperatures over the last two billion years. Instead, it shows that Earth has had a relatively stable and mild climate.

Temperature is an important control over chemical reactions that govern life and our environment. This ground-breaking work will have significant implications for scientists working on or questions surrounding biological and climate .

[…]

In the work, Dr. Isson and Ph.D. student Sofia Rauzi adopted novel methods to illuminate a history of Earth’s surface .

They utilized five unique data records derived from different rock types including shale, iron oxide, carbonate, silica, and phosphate. Collectively, these ‘geochemical’ records comprise over 30,000 that span Earth’s multi-billion-year history.

To date, the study is the most comprehensive collation and interpretation of one of the oldest geochemical records—. Oxygen isotopes are different forms of the element oxygen. It is also the first study to use all five existing records to chart a consistent ‘map’ of temperature across an enormous portion of geological time.

“By pairing oxygen isotope records from different minerals, we have been able to reconcile a unified history of temperature on Earth that is consistent across all five records, and the oxygen isotopic composition of seawater,” says Dr. Isson.

The study disproves ideas that early oceans were hot with temperatures greater than 60°C prior to approximately half a billion years ago, before the rise of animals and land plants. The data indicates relatively stable and temperate early-ocean and temperatures of around 10°C which upends current thinking about the environment that complex life evolved in.

The work produces the first ever record of the evolution of terrestrial (land-based) and marine clay abundance throughout Earth history. This is the first direct evidence for an intimate link between the evolution of plants, marine creatures that make skeletons and shells out of silica (siliceous life forms), clay formation, and .

“The results suggest that the process of clay formation may have played a key role in regulating climate on early Earth and sustaining the temperate conditions that allowed for the evolution and proliferation of life on Earth,” says Dr. Isson.

[…]

The work produces the first ever record of the evolution of terrestrial (land-based) and marine clay abundance throughout Earth history. This is the first direct evidence for an intimate link between the evolution of plants, marine creatures that make skeletons and shells out of silica (siliceous life forms), clay formation, and .

“The results suggest that the process of clay formation may have played a key role in regulating climate on early Earth and sustaining the temperate conditions that allowed for the evolution and proliferation of life on Earth,” says Dr. Isson.

Source: New evidence changes key ideas about Earth’s climate history

23andMe Thinks ‘Mining’ Your DNA Data Is Its Last Hope

Posted on by

23andMe is in a death spiral. Almost everyone who wants a DNA test already bought one, a nightmare data breach ruined the company’s reputation, and 23andMe’s stock is so close to worthless it might get kicked off the Nasdaq. CEO Anne Wojcicki is on a crisis tour, promising investors the company isn’t going out of business because she has a new plan: 23andMe is going to double down on mining your DNA data and selling it to pharmaceutical companies.

“We now have the ability to mine the dataset for ourselves, as well as to partner with other groups,” Wojcicki said in an interview with Wired. “It’s a real resource that we could apply to a number of different organizations for their own drug discovery.”

That’s been part of the plan since day one, but now it looks like it’s going to happen on a much larger scale. 23andMe has always coerced its customers into giving the company consent to share their DNA for “research,” a friendlier way of saying “giving it to pharmaceutical companies.” The company enjoyed an exclusive partnership with pharmaceutical giant GlaxoSmithKline, but apparently the drug maker already sucked the value out of your DNA, and that deal is running out. Now, 23andMe is looking for new companies who want to take a look at your genes.

[…]

the most exciting opportunity for “improvements” is that 23andMe and the pharmaceutical industry get to develop new drugs. There’s a tinge of irony here. Any discoveries that 23andMe makes come from studying DNA samples that you paid the company to collect.

[…]

The problem with 23andMe’s consumer-facing business is the company sells a product you only need once in a lifetime. Worse, the appeal of a DNA test for most people is the novelty of ancestry results, but if your brother already paid for a test, you already know the answers.

[…]

it’s spent years trying to brand itself as a healthcare service, and not just a $79 permission slip to tell people you’re Irish. In fact, the company thinks you should buy yourself a recurring annual subscription to something called 23andMe+ Total Health. It only costs $1,188 a year.

[…]

The secret is you just can’t learn a ton about your health from genetic screenings, aside from tests for specific diseases that doctors rarely order unless you have a family history.

[…]

What do you get with these subscriptions? It’s kind of vague. Depending on the package, they include a service that “helps you understand how genetics and lifestyle can impact your likelihood of developing certain conditions,” testing for rare genetic conditions, enhanced ancestry features, and more. Essentially, they’ll run genetic tests that you may not need. Then, they may or may not recommend that you talk to a doctor, because they can’t offer you actual medical care.

You could also skip the middleman and start with a normal conversation with your doctor, who will order genetic tests if you need them and bill your insurance company

[…]

If 23andMe company survives, the first step is going to be deals that give more companies access to look at your genetics than ever before. But if 23andMe goes out of business, it’ll get purchased or sold off for parts, which means other companies will get a look at your data anyway.

Source: 23andMe Admits ‘Mining’ Your DNA Data Is Its Last Hope

What this piece misses is the danger of whom the data is sold to – or if it is leaked (which it was). Insurance companies may refuse to insure you. Your DNA may be faked. Your unique and unchangeable identity – and those of your family – has been stolen.

US judge dismisses authors’ ridiculous copyright claim against OpenAI

Posted on by

A US judge has dismissed some of the claims made by writers in a copyright infringement lawsuit against OpenAI, though gave the wordsmiths another chance to amend their complaint.

The case – Paul Tremblay et al vs OpenAI – kicked off in 2023 when novelists Paul Tremblay, Christopher Golden, and Richard Kadrey, and writer-comedian-actress Sarah Silverman accused OpenAI of illegally scraping their work without consent to train the AI champion’s large language models.

The creators claimed that ChatGPT produced accurate summaries of their books and offered that as evidence that their writing had been ripped off. Since OpenAI’s neural networks learn to generate text from its training data, the group argued that its output should be considered a “derivative work” of their IP.

The plaintiffs also alleged that OpenAI’s model deliberately omitted so-called copyright management information, or CMI – think books’ ISBN numbers and authors’ names – when it produced output based on their works. They also accused the startup of unfair competition, negligence, and unjust enrichment.

All in all, the writers are upset that, as alleged, OpenAI not only used copyrighted work without permission and recompense to train its models, its model generates prose that closely apes their own, which one might say would hinder their ability to profit from that work.

Federal district Judge Araceli Martínez-Olguín, sitting in northern California, was asked by OpenAI to dismiss the authors’ claims in August.

In a fresh order [PDF] released on Monday, Martínez-Olguín delivered the bad news for the scribes.

“Plaintiffs fail to explain what the outputs entail or allege that any particular output is substantially similar – or similar at all – to their books. Accordingly, the court dismisses the vicarious copyright infringement claim,” she wrote. She also opined that the authors couldn’t prove that CMI had been stripped from the training data or that its absence indicated an intent to hide any copyright infringement.

Claims of unlawful business practices, fraudulent conduct, negligence, and unjust enrichment were similarly dismissed.

The judge did allow a claim of unfair business practices to proceed.

“Assuming the truth of plaintiffs’ allegations – that defendants used plaintiffs’ copyrighted works to train their language models for commercial profit – the court concludes that defendants’ conduct may constitute an unfair practice,” Martínez-Olguín wrote.

Although this case against OpenAI has been narrowed, it clearly isn’t over yet. The plaintiffs have been given another opportunity to amend their initial arguments alleging violation of copyright by filing a fresh complaint before March 13.

The Register has asked OpenAI and a lawyer representing the plaintiffs for comment. We’ll let you know if they have anything worth saying. ®

Source: US judge dismisses authors’ copyright claim against OpenAI • The Register

See also: A Bunch Of Authors Sue OpenAI Claiming Copyright Infringement, Because They Don’t Understand Copyright

and: OpenAI disputes authors’ claims that every ChatGPT response is a derivative work, it’s transformative