A former researcher at leading artificial intelligence firms has warned that staff within the industry are “genuinely frightened” by the pace of technological development, claiming there is a significant risk to human existence within the next decade.
Jacob Coxon recently resigned from the AI laboratory Anthropic, accusing the sector of “gambling with our lives.” Speaking in a series of broadcast interviews following his resignation on Wednesday, 9 September 2026, Mr Coxon told the BBC that AI staff are “genuinely frightened” for humanity’s future.
The warning coincides with an unusual call for restraint from within the corporate leadership of the AI industry. On Saturday, 12 September 2026, Anthropic CEO Dario Amodei called for a global slowdown in development. Mr Amodei cited “exponential” growth in the capabilities of new models as a clear warning sign that safety protocols may not be keeping pace with innovation.
National security concerns
In the House of Commons this week, Prime Minister Andy Burnham acknowledged that the government is monitoring the risks AI poses to national security. According to reports on the parliamentary session, the Prime Minister confirmed that the risks are being treated as a priority for the Cabinet Office.
The proposal for a ‘global slowdown’
Anthropic has proposed a “three-point plan” for what it calls “pacing the frontier.” This strategy includes the introduction of independent third-party evaluators to audit model safety.
The debate over regulation comes at a critical time for the tech sector. While researchers like Mr Coxon and Evan Hubinger—who also estimated a greater than 10 per cent extinction risk—continue to call for immediate intervention, the economic pressure to maintain current development speeds remains high.
