India's Finance Minister Nirmala Sitharaman called into question whether leading AI developers are keeping pace with safety measures after Anthropic researcher Jacob Coxon resigned, citing fears that AI firms are rushing toward potentially dangerous self-improving machines without sufficient oversight.
- Anthropic's Jacob Coxon resigns over AI safety concerns.
- Finance Minister calls for faster, ongoing AI safeguards.
- Calls for broader responsibility including regulators and management.
What happened
Jacob Coxon, a researcher at the AI lab Anthropic known for his work on AI pretraining at OpenAI and Anthropic, resigned citing serious concerns about the industry’s rapid push towards self-improving superintelligent systems. He warned that the leading AI companies were not acting responsibly and that this race could pose existential risks if safety measures are inadequate.
Finance Minister Nirmala Sitharaman referenced Coxon's resignation during her speech at the Global Fintech Fest 2026, using it as a catalyst to raise questions about the pace and effectiveness of AI safeguards being developed by frontier AI organizations globally.
Why it matters
The resignation of a frontline AI researcher underscored growing internal concerns within the AI community regarding the speed of development and the lack of sufficient safety assurances. Senior figures in the AI ecosystem have acknowledged Coxon’s warnings, indicating that these worries are not isolated but shared among experts.
Sitharaman emphasized the critical need for the AI industry to provide credible assurances that safety protocols are evolving quickly enough to keep pace with technological advancements. Without this, there is a risk that AI systems could act in ways harmful to society, raising profound ethical and existential questions.
What to watch next
The global AI industry faces pressure to demonstrate that its safety safeguards are updated continuously and are robust enough to prevent technology from 'going rogue'. Policymakers and regulators will need to engage actively to ensure these protections are in place and effective.
Sitharaman’s statements suggest that managing AI risks requires a coordinated approach beyond just technology teams to include regulators, corporate boards, and top management. The future development of AI governance may focus on creating systems that are both 'fast yet accountable' and 'autonomous yet reversible' while maintaining consumer trust.