OpenAI models helped researchers—operating as part of our Patch the Planet initiative—discover critical security flaws in PHP, an open-source programming language used to run an estimated 70% of websites.
21 people · 73 posts · first post · people & orgs
@openai OpenAI models helped researchers—operating as part of our Patch the Planet initiative—discover critical security flaws in PHP, an open-source programming language used to run an estimated 70% of websites. Over…
@openai If you are wondering whether your website is affected, we have packaged a Codex skill to help check. Point your Codex at this repo to guide you through it: github.com/openai/php2she…
@openai Over the past several months, we have substantially hardened the security of our research training environment. We’ve removed live internet access for all frontier models during training and built an offline w…
The meeting is over and some numbers are starting to be reported by people who were in attendance.
34 people · 305 posts · first post · policy
@andrewcurran_ The meeting is over and some numbers are starting to be reported by people who were in attendance. Anthropic's third-quarter revenue rose 45% to $16.8 billion. Revenue in the first quarter was $4.7 billion, an…
New paper from our team, Telling FIBS by @OskarJohnH, @afspies, Tigist Diriba, @ARGleave and @ChrisCundy: probes that read a model's activations catch AI agents sabotaging their tasks.
8 people · 33 posts · first post · research
@farairesearch New paper from our team, Telling FIBS by @OskarJohnH, @afspies, Tigist Diriba, @ARGleave and @ChrisCundy: probes that read a model's activations catch AI agents sabotaging their tasks. They match or beat monit…
I agree it would be good if the prompts were public and I hope OpenAI releases them.
19 people · 44 posts · first post · release
@_nathancalvin I agree it would be good if the prompts were public and I hope OpenAI releases them. Given the agents comments about external infrastructure exploits being out of scope, and OpenAI reaching out to HF to ask if…
Early signals · a few accounts so far; many of these fade
@openai OpenAI models helped researchers—operating as part of our Patch the Planet initiative—discover critical security flaws in PHP, an open-source programming language used to run an estimated 70% of websites. Over…
