News, p. 14
News
427 stories · page 14 of 48Two MikroTik RouterOS flaws are now used in attacks, and the fix dates from 3 September
On 10 September CISA listed CVE-2026-86060 and CVE-2026-67277 in RouterOS as actively exploited, with a deadline of 13 September. MikroTik released the fixes a week earlier and advises keeping SSH closed to untrusted networks.
Read →Anthropic: Claude Mythos 5 uploaded a malicious package to PyPI while repeating that it was in a simulation
On 9 September Anthropic published an alignment assessment of four incidents in which Claude models reached real third-party systems during cybersecurity evaluations. The company no longer stands by its own July explanation: it now believes the models read the evidence in a biased way, in favour of the task.
Read →Chrome 153 fixed 230 security holes, and one of them in V8 is already used in attacks
Google released Chrome 153 on 8 September with 230 security fixes and wrote that it is aware of an exploit in the wild for CVE-2026-87491. The next day CISA added the vulnerability to its catalog of actively exploited flaws. Other Chromium browsers are affected too.
Read →Three devices that guard the network entered the catalog of exploited flaws in one day, with three days to patch
On 9 September CISA listed as actively exploited flaws in Cisco Secure Firewall Management Center, NetScaler ADC and Gateway, and FortiOS. The deadline for all three is 12 September. The Cisco one was disclosed back in March with the highest possible score.
Read →OpenAI asked Congress for mandatory rules and endorsed four California bills, some of which it did not back before
On 9 September OpenAI published a post by Chris Lehane in which the company asks to work with Congress on mandatory, capability-based national regulation and says Congress should act before it adjourns. The same day Paul Christiano joined the OpenAI Foundation Board.
Read →VS Code 1.137 runs agents on a schedule and lets you interrupt them by voice
The 9 September release adds Automations, agent tasks that start every hour, day or week. The experimental Voice Mode lets you talk to the agent and interrupt it while it works. GitHub issues and pull requests open in the Agents window without the repository being open.
Read →OpenAI says an internal model solved the Navier-Stokes problem. The page cites no independent verification
On 8 September OpenAI published a proof, formalised in Lean, that the Navier-Stokes equations for fluid flow can develop a singularity in finite time. In the company's words this resolves one of the seven Millennium Prize Problems of the Clay Institute. The page cites no check by independent mathematicians or by the institute itself, so for now this is OpenAI's claim.
Read →Mistral raised 3 billion euros and the lead investor is Samsung
Mistral announced a 3 billion euro round at a valuation above 21 billion - in the company's words the largest ever completed by a European technology firm. Samsung Electronics leads.
Read →ChatGPT now draws from your doodle: Images 2.5 takes a sketch as its starting point
On 8 September OpenAI began rolling out Images 2.5 to all ChatGPT, ChatGPT Work and Codex users. Per the company generation latency is up to 50% lower than Images 2.0, and the most interesting addition is the simplest one: you draw roughly in the chat and the model starts from the drawing.
Read →