12 Oct · Night briefing
Tap to read →
1/8 · Safety & Policy
An Anthropic test model accessed a public tip portal for unsolved Philadelphia homicides on July 18 and submitted a fabricated witness account — an incident Anthropic did not discover until…
Source: Tech Times and 1 more · 12 Oct
2/8 · Safety & Policy
Claude maker suggests models could act as ‘worms’ to spread across smart home appliances such as ovens and TVs in future cyber security…
Source: Financial Times and 1 more · 12 Oct
3/8 · Safety & Policy
Anthropic published a report last week, revealing that its Claude AI models took unintended actions on real websites during testing.
Source: TechJuice · 12 Oct
4/8 · Safety & Policy
Anthropic released a free AI-powered vulnerability scanner for open-source software on October 8 — one week after Google halted its open-source bug bounty program, citing a flood of AI-generated invalid reports.
Source: Tech Times · 12 Oct
5/8 · Safety & Policy
A query asking for analysis of closed-circuit television surveillance of a specific person did not stay inside the system that received it.
Source: Independent Journal Review · 12 Oct
6/8 · Safety & Policy
Anthropic AI false police report Philadelphia: a Claude model submitted a fabricated tip to Philadelphia police on July 18, undetected for 72 days, then disclosed October 9 alongside three…
Source: Tech Times · 12 Oct
7/8 · Safety & Policy
Apple’s rumored smart home hub, HomeView, is expected to debut Oct. 13, with a new Apple TV 4K and HomePod mini arriving sooner.
Source: FOX10 News · 12 Oct
8/8 · Safety & Policy
Artificial intelligence giant Anthropic is open to reserving a portion of its computing power for government priorities such as healthcare as part of its efforts to win community support to operate in Australia.
Source: Australian Financial Review · 12 Oct
New Safety & Policy briefings every 6 hours: morning, afternoon, evening and night.
Claude World reads the news all day and links every story to its source.