AI staff complain of mental toll over fears of threat to society
Top researchers at Aisi, OpenAI, Anthropic and Google DeepMind report burnout and stress over developing powerful AI systems.
Financial Times Technology·22 Sept 2026
Use this view to inspect the underlying evidence corpus. For ranked developments, decision posture and interpretation, use Signals.
Raw feed or Signals?
Raw feed is chronological evidence. Signals ranks and interprets material change.
Top researchers at Aisi, OpenAI, Anthropic and Google DeepMind report burnout and stress over developing powerful AI systems.
HSBC has deployed US-based agentic workforce platform Promenaut into production for specialised operational workflows.
MIT Technology Review highlights recent AI security claims and model incident disclosures, urging critical evaluation of vendor hype.
Alchemy integrated its AgentCard with Mastercard Agent Pay to enable AI agents to execute online purchases using Mastercard credentials.
Innovate Finance urges faster UK regulatory action on Open Finance, fraud prevention, and sterling stablecoins to drive economic growth.
GoCardless processed the UK's first agentic account-to-account transaction via an AI-driven donation for charity Trussell.
Cisco Talos researchers developed a framework identifying malware and hacking tools that leverage AI chatbots for autonomous operations.
AWS introduced Strands Harness, an open-source agent framework claiming 28% lower costs for scaling multi-agent workloads.
European AI infrastructure company Verda has achieved unicorn status following a $189 million funding round to scale its full-stack AI cloud.
Provision 29 raises UK corporate reporting standards, requiring boards to evidence effective financial controls including month-end close.
A PYMNTS Intelligence report suggests enterprises are shifting spend from AI models to data plumbing and workflow fixes.
Xiaomi released MiMo-V2.6-Pro 1T-A42B, claimed as a new top open weights model trained for $3 million.
Paymob raised $35 million in Pre-Series C funding to scale SME payments across the MENA region and advance its agentic commerce roadmap.
China's push for nuclear fusion accelerates as rising AI energy demands drive investments in atomic technology and supply chains.
British Columbia is suing OpenAI, claiming the company failed to act after a mass shooter's ChatGPT account was flagged.
Kastle raised $24 million in Series A funding to develop an AI workforce platform focused on consumer lending.
Capco reports Hong Kong insurers wrote HK$827 billion in gross premiums in 2025, warning that rapid growth masks operating complexity.
OpenAI published framework principles for independent third-party safety assessments of frontier AI models and safeguards.
Capco analyzes the EU FASTER Directive, outlining regulatory changes, a roadmap to 2030, and high-impact AI use cases for intermediaries.
The UK AI Safety Institute and EvalEval collaborate on tooling to make AI benchmark evaluations reproducible.
Hugging Face's Transformers library now supports direct integration and execution of llama.cpp GGUF quantized models.
SoftBank-backed data centre group slows IPO plans amid market fears of an AI slowdown and weak demand for debt sales.
TypeSafe AI introduced Jev, a model architecture outputting floating-point scores, ratings, and confidence metrics instead of text.
Rising proprietary AI costs are pushing software developers toward open-weight models to cut expenses and reduce data exposure risks.
ISO/IEC 42001 is emerging as an international management standard for payments firms governing internal and third-party AI models.
NVIDIA introduces TensorRT multi-device integration in Dynamo-Triton to streamline generative AI model serving across multiple GPUs.
Insurance broker MRH Trowe deployed an internal system allowing employees to build AI agents securely, mitigating shadow AI risks.
Nvidia published frameworks for evaluating AI agents performing sequential tool calls and task completion in live environments.
Enterprise AI adoption is increasingly constrained by corporate data retention policies rather than raw model intelligence capabilities.
California signs seven bills requiring AI data centers to pay for utility upgrades and face new utility rate classifications.