OpenAI announces the update of the GPT-5.5-Cyber model (new), which achieves a 85.6% score on the CyberGym benchmark compared to 81.9% in its initial version. Codex also obtained a new security plugin.
REGULATION
-
Democrats less confident in AI regulation than Republicans in 2026
By
–
Democrats are the more regulation skeptical ones now. In 2026, 74% of Democrat US adults say they have little to no confidence in the US goverment regulating AI effectively. Republicans are more confident in government regulation, at 61%. In 2024, those positions were flipped:
-
OpenAI demands US regulation, GPT Cyber New beats Claude Mythos on CyberGym
By
–
OpenAI is practically calling for regulation by the US government. No kidding: GPT Cyber New outperforms Claude Mythos on CyberGym.
-
AI Governance Problem: The Real Challenge for Companies
By
–
Most companies do not have an AI problem. They have an AI governance problem. As agents move across systems, data, and workflows, the real challenge becomes: Visibility Control Accountability Microsoft’s Agent 365 updates point to the next enterprise bottleneck:
-
Educational purposes excuse still bypasses AI guardrails
By
–
funny how “don t worry it's for educational purposes” still works against most AI guardrails
-
US government hampers expensive American AI models, industry seeks workaround with small models
By
–
The American models are the ones we want to use, but they're more expensive and now they're hamstrung by the U.S. government, which leaves this industry in a predicament. But you're right: we'll find a way around the damage. One way around it is to use a bunch of small models to
-
Claude Fable 5, Mythos 5 Offline After US Export Control Directive
By
–
3/ Claude Fable 5 and Mythos 5 are still offline. The US issued an export-control directive on Jun 12. Anthropic complied but publicly disagreed. Execs in Seoul said models could return "in coming days." No restoration notice from Anthropic yet. https://
buff.ly/LhrFXL4 -
Future AI models will be so powerful they self-ban
By
–
The future models will be so powerful that they will ban themselves.
-
AI Robotics Challenges in 2030s Impact Employment and Pensions
By
–
Additional challenges will be posed by advanced AI robotics in the 2030s and their impact upon employment and pensions funding.
-
ChatGPT record and replay unavailable in Europe
By
–
gdammit, chatgpt record and replay not avail in europa