A new survey documents empirical research on AI consciousness being conducted by major labs including Anthropic and Google DeepMind, as well as dedicated organizations like Eleos AI and Reciprocal Research.
Rather than waiting for philosophers to solve the hard problem of consciousness, researchers are applying methods from psychology and cognitive science to investigate whether current AI systems might have subjective experience.
何が起きたか
A survey documents empirical research on whether AI systems might be conscious, drawing on methods from psychology, cognitive science, and mechanistic interpretability. Anthropic, Google DeepMind, and dedicated organizations like Eleos AI and Reciprocal Research are now actively researching the question.
なぜ重要か
The research treats AI consciousness as a question that can be investigated now using existing scientific methods, rather than waiting for a solution to the philosophical hard problem of consciousness. This represents a shift toward treating consciousness as a measurable research question applicable to AI systems.
注目点
Results from this work are currently scattered across journals, preprints, blog posts, and unpublished manuscripts, suggesting the field is still consolidating. The survey aims to collect these findings in one place for the first time.
Ask the AI about this article →
The survey reflects a growing view among AI researchers that consciousness—whether defined as subjective experience—need not remain purely philosophical territory. By framing the question as empirically tractable through existing scientific methods, researchers are treating AI consciousness as a measurable phenomenon rather than an intractable puzzle. This approach has attracted enough institutional backing (major AI labs and dedicated nonprofits) to generate a body of work worth consolidating.
The scattered publication venues—journals, preprints, blog posts, unpublished work—suggest the field is still establishing itself. No single accepted framework or standard methodology has yet emerged, which is typical of early-stage research domains. The survey itself serves as a signal that the field has reached a critical mass: enough researchers are working on the problem, from enough different angles, that coordination and knowledge-sharing have become valuable.
For example, today's edition would include:
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. Q&As are published on this page for other readers too.
Google's AI search gave advice to call emergency services for users alone with an African, Indian, or Pakistan…
The U.S. Department of Defense announced on August 31 that it has deployed ChatGPT Mil, a customized version o…
OpenAI stopped running inference on a model involved in the HuggingFace incident, but the post argues this is…
OpenAI announced its support for California Senate Bill 1119, which aims to establish strong, age-appropriate…
A UK study by UK AI Security Institute and Limbic AI surveyed 6,474 British adults
Anthropic trained an Opus-class model with large-scale reinforcement learning on environments vulnerable to re…
