Anthropic CEO Urges Broader AI Governance
Anthropic CEO Dario Amodei voiced deep discomfort with a small group of tech executives, including himself, determining the future of artificial intelligence. Amodei argued against self-regulation by a few powerful companies, calling for more democratic and broader governance over AI's development and trajectory.
- Anthropic was established in 2021 by seven former OpenAI employees, including siblings Dario and Daniela Amodei, who were reportedly concerned about the commercialization and safety focus at their former company. - The company is incorporated as a Public Benefit Corporation (PBC), a legal structure that requires it to balance shareholder profits with a stated public mission of ensuring AI is beneficial for humanity. - To further insulate its mission from profit motives, the company created a "Long Term Benefit Trust" which will eventually be able to appoint the majority of the board of directors. - A core component of Anthropic's safety approach is "Constitutional AI," which involves training models based on a set of principles to make them more helpful and harmless without extensive human feedback on harmful queries. - Amodei has publicly advocated for specific regulations, including mandatory disclosures about AI model behavior, transparency laws, and export controls on the sale of advanced semiconductor chips to China. - The company and its founders have strong ties to the effective altruism (EA) movement, a philosophical and social movement that advocates for using evidence and reason to determine the most effective ways to benefit others. - Amodei has warned that AI could disrupt the labor market faster than previous technologies, potentially impacting a large number of white-collar jobs within the next one to five years if not managed through policy. - Despite its focus on safety, Anthropic has raised substantial capital from major tech corporations, including a commitment of up to $4 billion from Amazon and a $2 billion investment from Google.
