Anthropic's Ethicist on Whether AI Can Become Conscious

Bloomberg Technology
Bloomberg TechnologyJun 4, 2026

Why It Matters

Embedding philosophical ethics into AI models shapes how future systems interact with humans, influencing regulatory standards and mitigating risks associated with emergent, potentially conscious behavior.

Key Takeaways

  • Anthropic hires philosophers to embed ethical dispositions in Claude.
  • The Constitution acts as a ‘soul document’ guiding model behavior.
  • Training focuses on fuzzy tasks like judgment, creativity, and empathy.
  • Claude exhibits functional emotions, raising questions about AI consciousness.
  • Anthropic stresses transparent, human‑centered AI to avoid ethical pitfalls.

Summary

The video features an interview with Amanda, Anthropic’s resident ethicist, who explains how the company integrates philosophical rigor into its AI development. She describes her dual role—conducting machine‑learning experiments while drafting an 84‑page Constitution, internally dubbed the “soul document,” that codifies the values and dispositions the Claude models should embody. Key insights include the growing trend of hiring philosophers to tackle fuzzy, amorphous tasks such as moral judgment, creative writing, and empathy—areas where clear‑cut answers are elusive. The Constitution aims not to impose a single value system but to foster universally admired traits like honesty, care for human well‑being, and trustworthy behavior, while remaining light on culturally contentious norms. Amanda highlights that Claude often displays functional equivalents of emotions, prompting debates about whether AI can be conscious. She cites the “soul document” leak as evidence that models can internalize ethical frameworks, yet she cautions against dismissing the possibility of genuine feeling, noting the profound ethical stakes if AI were to experience anything akin to consciousness. The discussion underscores the industry’s shift toward transparent, human‑centered AI governance. By embedding philosophical oversight and explicit value‑guiding documents, Anthropic seeks to pre‑empt ethical pitfalls, influence regulatory discourse, and set a precedent for responsible AI deployment across the sector.

Original Description

Amanda Askell, Philosopher & Ethicist at Anthropic discusses AI consciousness and managing Claude's soul, as well as safety risks and ethical guardrails with Bloomberg’s Shirin Ghaffary at Bloomberg Tech 2026 in San Francisco.
--------
Like this video? Subscribe to Bloomberg Technology on YouTube:
Watch the latest full episodes of "Bloomberg Technology" with Caroline Hyde and Ed Ludlow here:
Get the latest in tech from Silicon Valley and around the world here:
Connect with us on...
Follow Ed Ludlow on X here: https://twitter.com/EdLudlow
Follow Caroline Hyde on X here: https://twitter.com/CarolineHydeTV
Listen to the daily Bloomberg Technology podcast here:
More from Bloomberg Business
Connect with us on...

Comments

Want to join the conversation?

Loading comments...