Home

a shaneful place

News

Artificial Intelligence

Blog

Portfolio

Subscribe

AI News Digest - 2026-09-19

Category

Empty

1.

California Governor Gavin Newsom signed an executive order that required independent auditors inside AI labs, called for a "kill switch" for AI models, and established an expert panel to deliver recommendations within two months.

2.

Security researchers used Anthropic's Claude models to breach OpenAI's internal systems through its community forum in under 72 hours, demonstrating that newer AI models could reduce the time and expertise required to exploit security vulnerabilities.

3.

OpenAI and Microsoft faced internal disclosures in which employees described training-data practices as "astonishing theft" and "largely substitutive," statements that undercut the companies' reliance on fair use defenses.

4.

MIT Technology Review hosted a live roundtable on catastrophic AI risk and reported that alignment remained unsolved, monitoring tools were fragile, and meaningful US federal action on AI regulation had not occurred.

5.

Google DeepMind said that visible chains of thought were a safety advantage for AI but warned that such transparency was slipping away.

References

1.

https://the-decoder.com/california-governor-newsom-signs-executive-order-demanding-kill-switch-for-ai-models/

California Governor Newsom signs executive order demanding "kill switch" for AI models

California Governor Gavin Newsom signed an executive order seeking independent auditors inside AI labs and a "kill switch" for AI models. An expert panel has two months to deliver recommendations. Newsom says no federal law requires AI companies to report dangerous incidents.

the-decoder.com

1.

https://the-decoder.com/security-researchers-used-anthropics-claude-to-hack-openais-internal-systems-in-under-72-hours/

Security researchers used Anthropic's Claude to hack OpenAI's internal systems in under 72 hours

Three security researchers used Anthropic's Claude models to break into OpenAI's internal systems through its community forum in less than 72 hours. According to the team, Opus 5 succeeded where its predecessor couldn't bypass a common security measure. The attack shows how newer AI models can cut the time and expertise needed to exploit security flaws.

the-decoder.com

1.

https://the-decoder.com/ai-training-built-on-fair-use-looks-shaky-when-the-companies-own-people-call-it-astonishing-theft/

AI training built on fair use looks shaky when the companies' own people call it "astonishing theft"

Internal emails and sworn testimony undercut OpenAI and Microsoft's fair use defense. A Microsoft director described the practice as the "largest theft of labor in human history," while OpenAI's head of ChatGPT wrote that the products "are largely substitutive, period."

the-decoder.com

1.

https://www.technologyreview.com/2026/09/18/1144435/could-ai-really-kill-us-all-your-questions-answered/

Could AI really kill us all? Your questions, answered.

You had a lot of questions about the recent claims that humanity is facing an existential risk from AI. We had a go at answering a few.

technologyreview.com

1.

https://the-decoder.com/visible-chains-of-thought-are-a-safety-advantage-for-ai-but-that-transparency-is-slipping-away/

Visible chains of thought are a safety advantage for AI, but that transparency is slipping away

AI models think out loud today, but Google Deepmind says that transparency is at risk.

the-decoder.com

Made with Slashpage