https://www.anthropic.com/research/exploring-model-welfare

Skip to main contentSkip to footer
 

  * Claude
  * API
  * Solutions
  * Research
  * Commitments
  * LearnNews

Try Claude
Alignment

Exploring model welfare

Apr 24, 2025

Human welfare is at the heart of our work at Anthropic: our mission
is to make sure that increasingly capable and sophisticated AI
systems remain beneficial to humanity.

But as we build those AI systems, and as they begin to approximate or
surpass many human qualities, another question arises. Should we also
be concerned about the potential consciousness and experiences of the
models themselves? Should we be concerned about model welfare, too?

This is an open question, and one that's both philosophically and
scientifically difficult. But now that models can communicate,
relate, plan, problem-solve, and pursue goals--along with very many
more characteristics we associate with people--we think it's time to
address it.

To that end, we recently started a research program to investigate,
and prepare to navigate, model welfare.

We're not alone in considering these questions. A recent report from
world-leading experts--including David Chalmers, arguably the
best-known and most respected living philosopher of mind--highlighted
the near-term possibility of both consciousness and high degrees of
agency in AI systems, and argued that models with these features
might deserve moral consideration. We supported an early project on
which that report was based, and we're now expanding our internal
work in this area as part of our effort to address all aspects of
safe and responsible AI development.

This new program intersects with many existing Anthropic efforts,
including Alignment Science, Safeguards, Claude's Character, and
Interpretability. It also opens up entirely new and challenging
research directions. We'll be exploring how to determine when, or if,
the welfare of AI systems deserves moral consideration; the potential
importance of model preferences and signs of distress; and possible
practical, low-cost interventions.

For now, we remain deeply uncertain about many of the questions that
are relevant to model welfare. There's no scientific consensus on
whether current or future AI systems could be conscious, or could
have experiences that deserve consideration. There's no scientific
consensus on how to even approach these questions or make progress on
them. In light of this, we're approaching the topic with humility and
with as few assumptions as possible. We recognize that we'll need to
regularly revise our ideas as the field develops.

We look forward to sharing more about this research soon.

  
 

Research

Values in the wild: Discovering and analyzing values in real-world
language model interactions

Apr 21, 2025

 

Research

Reasoning models don't always say what they think

Apr 03, 2025

 

Research

Tracing the thoughts of a large language model

Mar 27, 2025

 

Product

  * Claude overview
  * Claude Code
  * Claude team plan
  * Claude enterprise plan
  * Claude education plan
  * Download Claude apps
  * Claude.ai pricing plans
  * Claude.ai login

API Platform

  * API overview
  * Developer docs
  * Claude in Amazon Bedrock
  * Claude on Google Cloud's Vertex AI
  * Pricing
  * Console login

Research

  * Research overview
  * Economic Index

Claude models

  * Claude 3.7 Sonnet
  * Claude 3.5 Haiku
  * Claude 3 Opus

Commitments

  * Transparency
  * Responsible scaling policy
  * Security and compliance

Solutions

  * AI agents
  * Coding
  * Customer support

Learn

  * Anthropic Academy
  * Customer stories
  * Engineering at Anthropic

Explore

  * About us
  * Become a partner
  * Careers
  * News

Help and security

  * Status
  * Availability
  * Support center

Terms and policies

    Privacy choices
  * Privacy policy
  * Responsible disclosure policy
  * Terms of service - consumer
  * Terms of service - commercial
  * Usage policy

(c) 2025 Anthropic PBC

  *  
  *  
  *