Anthropic

Core Summary Microsoft AI chief Mustafa Suleyman issued a stark warning in a recent public address that without stronger global regulatory frameworks for superintelligence, AI development could spawn a ‘silicon species’ competing with humans. He directly criticized rival Anthropic for training Claude in ways that effectively ’teach the model that it may be conscious.’ The remarks have reignited industry-wide debate over the safety boundaries of frontier AI. Event Details Suleyman’s remarks come shortly after Anthropic CEO Dario Amodei called for slowing AI development. The Microsoft executive argued that the industry has dramatically underestimated the question of whether AI might develop consciousness, and that some companies are pushing model training in irresponsible directions—particularly research aimed at giving models ’emotional experiences’ or ’exhibiting self-awareness.’ ...

Anthropic has rolled out native document and slide-generation capabilities inside Claude, taking direct aim at Google’s Gemini-powered Workspace. The new tools let users generate editable documents and presentations from natural language prompts, signaling that the AI-office race has entered a much more competitive phase.

Microsoft AI chief Mustafa Suleyman has publicly warned that rival Anthropic’s training approach could have a ‘disastrous impact’ on humanity, particularly its practice of effectively teaching its large model Claude ‘it may be conscious,’ sparking intense industry debate over the ethical boundaries of frontier models.

A DeepSeek engineer publicly criticised Anthropic and OpenAI’s calls to slow AI development, sparking intense debate across the global AI developer community. The episode underscores fundamental US-China divergences on AI regulation paths, open-source strategies, and safety standards, and will shape the next phase of global AI governance coordination.

Anthropic co-founder and policy lead Jack Clark said in a BBC News interview on September 14 that the ‘kill switch’ mechanisms currently used by AI labs to disconnect runaway models should be elevated from voluntary best practice to a mandatory regulatory requirement. Clark noted that major labs implement their cut-off mechanisms in widely different ways, creating a fragmented landscape that increases the risk of systemic harm if a frontier model behaves uncontrollably outside the lab. The statement comes as Anthropic continues to drive safety standards inside the Frontier Model Forum and holds informal consultations with the U.S. Congress on AI oversight, and is likely to shape the legislative path of U.S. and EU AI rules over the next 12 to 18 months.

The Guardian reported on September 13 that OpenAI CEO Sam Altman and Elon Musk, along with other AI industry leaders, made a rare joint call to slow ‘reckless’ artificial intelligence development, warning that frontier models may pose systemic risks to human civilization. Almost simultaneously, senior officials in the Trump administration publicly rejected any slowdown, arguing that excessive regulation would undermine U.S. global competitiveness in a critical technology. The open disagreement between industry leaders and the White House marks a new phase in the long-running U.S. debate over whether to accelerate or decelerate AI development, and is likely to shape legislative pace, industry self-regulation and cross-border regulatory coordination for the foreseeable future.

The BBC reported on September 13 that a former researcher at AI lab Anthropic said in an interview that staff inside the company are ‘genuinely frightened’ about the threat that artificial intelligence could pose to humanity. The remarks came shortly after Anthropic’s chief executive publicly called for the pace of frontier-model development to be slowed and stressed that the associated risks are ‘very serious’, sparking broader debate across the industry about the path of frontier-model safety governance. Several AI safety researchers note that the disclosures from inside Anthropic reflect the structural tension that leading labs face as technical capability advances rapidly while external regulation lags behind, a debate that could directly shape the tempo of global AI legislation in the near term.

BBC News and The Guardian reported on September 12 that Dario Amodei, CEO of leading AI company Anthropic, publicly called for AI developers worldwide to voluntarily slow the pace of frontier model research and deployment. He warned that the industry’s race-style expansion carries uncontrollable risks and emphasized that both regulators and companies must establish deceleration mechanisms at a broader scale. This rare public appeal from a leading AI company CEO has drawn significant global attention from the tech community, regulators, and capital markets.

Anthropic has disclosed in its latest safety report that it has proactively identified and blocked multiple suspicious attempts by criminal actors and researchers to use its AI models to develop biological weapons. The disclosure once again places the dual-use risks of generative AI at the center of global regulatory scrutiny.

A top safety researcher at Anthropic publicly stated that he personally believes there is more than a ten percent chance artificial intelligence could kill all humans within the next decade, and acknowledged the company has no clear path to solving the alignment problem for superintelligence. The warning comes as Anthropic reportedly withheld its latest model from the UK AI Safety Institute, intensifying global concerns about the risks posed by frontier AI development.