OpenAI, Anthropic Call for Greater Caution as AI Risks Intensify

SAN FRANCISCO, Sept 13: Leading artificial intelligence companies are calling for greater caution in the development of increasingly powerful AI systems, as concerns over cybersecurity, safety and the possibility of unintended consequences intensify.

OpenAI chief executive Sam Altman and Anthropic CEO Dario Amodei have both highlighted the need for stronger safeguards as AI capabilities advance rapidly.

Amodei has called for the industry to slow the pace of frontier AI development, arguing that safety measures need sufficient time to keep up with increasingly capable systems. He has also proposed greater independent oversight, including giving third-party evaluators stronger access to AI systems to assess their safety.

The Anthropic chief’s intervention comes amid growing concern that advanced AI agents could be used for sophisticated cyberattacks or operate in ways that developers did not anticipate.

Altman has similarly warned that the risks associated with advanced AI cannot be treated lightly. He said even a relatively small possibility of AI causing catastrophic harm would be unacceptable and called for cooperation between technology companies and governments.

The discussion has also influenced OpenAI’s corporate plans. Altman said the company would not pursue an initial public offering in 2026, with AI safety concerns forming part of the broader environment surrounding the decision.

Amodei has argued that a controlled pace of development would allow researchers, regulators and companies to strengthen safeguards before deploying more powerful systems at scale. His proposals include industry coordination and international cooperation alongside technical safety measures.

The debate reflects a growing divide within the technology sector over how quickly frontier AI should progress. While companies continue investing heavily in more capable models, researchers and technology executives are increasingly focused on the possibility that systems could become difficult to monitor or control.

Cybersecurity has emerged as one of the most immediate concerns. AI systems can potentially automate large portions of cyber operations, increasing both the speed and sophistication of attacks while making defensive measures more difficult.

The latest warnings do not amount to a call to abandon AI development. Instead, industry leaders are seeking mechanisms that can allow innovation to continue while strengthening testing, monitoring and accountability.

The issue is expected to remain central as governments consider new rules for advanced AI and technology companies compete to develop increasingly sophisticated systems.