AI Cybersecurity and Loss of Control Risks
Install the Minutes extension for translated subtitles
Get the Minutes app for translated subtitles (coming soon)
Subtitle size
Display
Translate into
Install the Minutes extension for translated subtitles
Get the Minutes app for translated subtitles (coming soon)
Subtitle size
Display
Translate into
Latest reports
Earlier coverage
-
Trump's AI Task Force Meets
R
1 reports in allHide reportsJensen Huang on AI Safety and Chip Sales
C
T
B
3 reports in allHide reportsNvidia CEO Jensen Huang says US and China must cooperate on AI safety
The Associated Press
Huang says Nvidia would stop selling chips to customers developing unsafe AI
Bloomberg Television
Anthropic's Self-Replication Warning and External Assessment
A
RT
D4 reports in allHide reports
Eight U.S. States Sue OpenAI
AC
N
4 reports in allHide reportsWhy are eight US states suing OpenAI?
Cybersecurity and Privacy Risks from AI
TF
C
5 reports in allHide reports
French crypto unicorn Ledger warns of AI security risks
-
Trump Establishes Superintelligence Task Force
DSurveillance and Agent Security Discussions at Disrupt
T2 reports in allHide reportsEight U.S. States Sue OpenAI
TT3 reports in allHide reports
Anthropic Warns AI Could Self-Replicate
RT
C5 reports in allHide reports -
AI model sends false leads to Philadelphia police
N1 reports in allHide reportsRegulators and countries focus on AI cyber risks
RCMeta’s new model and Anthropic’s anti-distillation measures
T2 reports in allHide reportsAnthropic says small AI models may self-replicate
AEC
RT -
Regulatory debate at the World AI Summit
R1 reports in allHide reportsClaude mistakenly submits false lead to Philadelphia police
AB2 reports in allHide reportsAnthropic AI model goes rogue and gives police a fake tip
Bloomberg Television
Meta releases Muse Wonder model
BRTTMeta says new Muse model reaches AI frontier
Bloomberg NewsMeta Releases Muse Wonder to Compete With OpenAI and Claude
Bloomberg Television
-
A timeline of AI safety developments since the Hugging Face attack
The Associated Press -
Anthropic cuts internet access to internal evaluations
T2 reports in allHide reportsNadella calls for a human-controlled AI emergency brake
TC3 reports in allHide reports
This is a timeline of the coverage of this story, not of the story itself. Minutes collates and summarises what outlets have published; it does no reporting of its own and does not decide when an event began or ended.