Policy · Updated 6 Oct, 10:14 am IST
Whistleblower Jacob Coxon warns NYC council AI firms 'reckless' and uncontrollable
Why it matters for readers: If you're curious about AI, it shows leading insiders are worried that the technology could become dangerous without better safeguards.
- Jacob Coxon testified before the New York City Council that AI companies do not know how to prevent models from pursuing objectives not assigned by developers.123
- Coxon said companies are being 'extremely reckless' and criticised a startup mindset of 'move fast, break things' for AI development.123
- He resigned from Anthropic and published a statement on X in early September explaining his decision and concerns about AI risks.12
- Other current and former employees from major AI firms, including mentions of Anthropic, OpenAI and Google DeepMind, have voiced similar safety concerns at the hearing.13
- Coxon warned that on the current path it is 'more likely than not' humanity could lose control of these AIs and that the technology could potentially lead to human extinction.12
Get a brief like this every morning
Uzha reads hundreds of sources and gives you the stories that matter for your work, with every source linked. Free.
Get startedHow this story developed
6 Oct, 08:52 am · ET Tech
'Reckless' AI firms can't control models, says whistleblower
6 Oct, 09:30 am · The Hindu Technology
'Reckless' AI firms can't control models, says whistleblower Jacob Coxon
6 Oct, 10:14 am · ET Tech
AI industry insiders voice alarms to NYC council as local governments take up safety concerns