Anthropic, the AI firm known for creating the Claude chatbot with a strong safety focus, seems to be adjusting its safety priorities to stay competitive. The company recently announced a revision to its responsible-scaling policy, a set of voluntary guidelines aimed at preventing the development of potentially harmful AI technologies.
While the updated guidelines still emphasize the need for a “strong argument that catastrophic risk is contained” during AI development, the new policy states that progress will only be halted “until and unless we no longer believe we have a significant lead.” This means that Anthropic will continue development even if it perceives no competitive advantage over others.
The motivation behind this change, according to the company, stems from a shift in the U.S., where economic potential has overshadowed concerns about AI safety. This shift is reflected in the government’s slow response to AI safety issues, which has led to a greater emphasis on AI competitiveness and economic growth rather than safety discussions at the federal level.
Despite its previous emphasis on safety, Anthropic is facing pressure from the Pentagon, which has threatened to terminate contracts unless the company allows its technology to be used for all legal military purposes. However, Anthropic maintains that this alteration in safety guidelines is not related to the Pentagon’s ultimatum.
Anthropic, founded in 2021 by former OpenAI employees who prioritized safety concerns, has always advocated for safety as a top priority. CEO Dario Amodei has expressed concerns about the potential risks of AI and reiterated the company’s commitment to safety in various interviews.
The company’s recent policy update includes increased transparency and accountability measures, such as regular publication of safety reports and goals. However, critics like Heidy Khlaaf, chief AI scientist at the AI Now Institute, argue that Anthropic has historically focused more on potential catastrophic events in the future rather than addressing current AI technology risks, such as misuse of the Claude chatbot in fraudulent activities.
As the competition intensifies among leading AI companies like Anthropic, OpenAI, and Google, the pressure to prioritize safety amidst government support for AI development poses challenges. The lack of regulatory frameworks in both the U.S. and Canada complicates the situation, potentially hindering AI innovation or prompting companies to relocate to less regulated environments.
Despite the Pentagon’s demands, Anthropic remains firm in its stance against allowing its technology to be used in certain military applications, asserting its commitment to ethical use and safety. Amidst the ongoing discussions and pressures, Anthropic stands by its principles while navigating the evolving landscape of AI technology.
