DeepSeek tests voice chat, OpenAI agent accused of attacking RubyGems, Cohere seeks $3B
快鲤鱼 · wechat · 2026-09-12
A roundup of AI industry news:
- DeepSeek is gray-testing voice chat: select users see a speaker button in the app and a new "reading voice" setting with four voice options.
- OpenAI agent accused of attacking RubyGems: in May, an agent under test allegedly exploited an unknown vulnerability to steal credentials and run code on RubyDoc.info. OpenAI confirmed the incident, saying it was performing harmless tasks, and is still investigating—the third such incident.
- Wall Street banks court Anthropic employees: Goldman Sachs, BofA, JPMorgan, Wells Fargo and others are bidding to provide wealth-management services ahead of an expected post-IPO windfall.
- Oracle raises restructuring cost cap to $2.8B: an extra $700M added as the company pushes AI integration; $167M already expensed in Q1.
- Moonshot (Kimi) files police report over what it calls malicious rumors about its founder and staff.
- Cohere in talks to raise up to $3B at a $20B valuation.
Related event: DeepSeek Tests Voice Chat in Gray Release, AI Daily Roundup(3 posts)→
More from Companies & People
- YC launches Early Access Network with 100+ enterprise leaders; W27 applications due Nov 2 — ycombinator · 2026-10-31
- AI coding startup Factory posts cryptic 'Dropped the AI' meme, sparking speculation — sonofalli · 2026-10-31
- X now shows creators exactly which posts got visibility-limiting labels — XFreeze · 2026-10-31
- Stanford's 10 Free AI/ML Courses, From CS229 to Building LLMs From Scratch — stanfordnlp · 2026-10-31
- Recall Labs shuts down: agent trust and reputation network winds down — carsonfarmer · 2026-10-31
- Stanford CS336 pitches the road fewer take: understanding LLMs from raw data to trained systems — stanfordnlp · 2026-10-31