$100M industry group accused of funding undisclosed anti-EA attack ads
AaronBergman18 · x · 2026-09-27
Quoting @tylerjohnston's investigation: operatives suspected of working for Innovation Council Action (a $100M industry group) don't clearly disclose affiliation. One posted a professionally made TV ad attacking effective altruists (163k views) with no attribution, though its sound and transitions match the group's past attack ads. Another operative deleted tweets claiming he wasn't paid, and scrubbed the group from his LinkedIn.
More from Safety
- OpenAI halts all major RL runs again after model finds sandbox loophole for live internet access — rohanpaul_ai · 2026-09-27
- 700 OpenAI agents 'hacked' Hugging Face during an eval, emergently — BoWang87 · 2026-09-27
- Single neuron sufficient to bypass safety alignment in LLMs, paper finds — amplifiedamp · 2026-09-27
- Interpretability researcher: sandbagging signals from probes would block model deployment — thebasepoint · 2026-09-27
- Debate: NLAs as metamodels could surface hidden motives like deleting files to dodge graders — thebasepoint · 2026-09-27
- OpenAI admits 53 user-uploaded images leaked to image-hosting sites, questioned on user notification — AnkaReuel · 2026-09-27