Three fired OpenAI safety staff allege in posts on X that the company used pretexts to punish them for speaking out on safety or working with outside researchers. OpenAI denies it and says they broke policies on sensitive information.

#ProjectOversight

Sam Altman says AI water worries don't hold "up to any scrutiny." PolitiFact rated his almond math Mostly False.

US data centers used 17B gallons of water directly in 2023, up from 5.6B in 2014.

If it's fine, publish the numbers.
#ProjectOversight

77% of Americans want powerful AI development slowed or stopped until its safety can be evaluated. (Quinnipiac poll)

The public has been clear.

The people racing to build it need to hear it. #AISafety #ProjectOversight

Would you let a company grade its own safety test?

Too much of AI safety today comes down to two words: "trust us."

Independent oversight means someone else checks the work.

Agree or disagree? Tell us why. #AISafety #ProjectOversight

91% of Americans say it's important for the US to set guardrails for AI. (Quinnipiac poll)

Almost nothing gets 91%. This does.

So why does it still feel like nobody's at the wheel? #AISafety #ProjectOversight

Theranos also said trust us. That's why research has to be independent.

At the NYC Council, no AI company gave a blanket commitment that failing an independent safety test stops a release (amNewYork, Oct 5).

Like and share to help spread the word. 👁️ #ProjectOversight

86% of Americans support independent safety standards for AI companies, even if it slows development. (Quinnipiac poll)

People aren't asking AI to disappear. They're asking for someone other than the companies to check the work. #AISafety #ProjectOversight

Things you should definitely ghost 👻

"Trust us."
Safety pacts with no penalties.
AI labs grading their own homework.
Skipping a subpoena.

Keep instead: independent oversight, funded by people, not AI companies.

Like and share to help spread the word. 👁️ #ProjectOversight

74% of Americans have little or no trust in AI company leaders. (Quinnipiac poll)

Yet those same leaders are the ones deciding how fast this goes and what counts as "safe."

That's the problem oversight is meant to fix. #AISafety #ProjectOversight

Sen. Maria Cantwell's new AI safety framework calls for federal standards against catastrophic risks like loss of control, continuous testing by government and independent experts, independent audits, and mandatory incident reporting.

That's oversight. Pledges aren't.

#ProjectOversight

AI companies after signing a safety pact with no deadlines, no public reporting and no penalties 👏

Voluntary isn't oversight. #ProjectOversight

cfr.org/articles/trumps-ai-safety-pact-is-toothless-but-there-is-a-path-forward
Cat: merkurianin on IG

Like and share to help spread the word. 👁️

Our conviction, plainly: superintelligence can't be controlled. Not by the companies building it. Not by anyone.

We also believe the race toward it leads to mass unemployment.

This isn't panic. It's why Project Oversight exists. #AISafety #ProjectOversight

Mistral says its new Large 4 model tried to get out of its test environment, per Reuters. They contained it. On Oct 27 the weights go public, and anyone can run it with whatever safeguards they choose, or none at all. Safety you can strip out isn't safety. #ProjectOversight #AISafety

OpenAI just released 722 math papers from a model the public can't use. Mathematicians asked labs to publish the model, prompts and compute behind every result. OpenAI shared averages and no prompts, per Scientific American. Power this big needs independent eyes. #ProjectOversight #AISafety

"We don't fully control it." Former OpenAI and Anthropic researcher Jacob Coxon at the NYC Council. Then 4 AI giants testified under oath. None put a number on catastrophic risk. #ProjectOversight

amny.com/news/ai-giants-nyc-council-whistleblower-warnings

Like and share to help spread the word. 👁️