Claude May Experience “Functional Emotions”

January 29, 2026

Ai,Chatbots,With,Virtual,Assistant,Tools,And,Icons,Representing,Business,

Silicon Valley elites are pushing society toward granting rights to machines, claiming their AI chatbots might have feelings and deserve emotional consideration. This dramatic shift is epitomized by companies like Anthropic, which released a 23,000-word “constitution” for its Claude chatbot, speculating that the software may experience “functional emotions” worthy of moral patienthood.

Story Snapshot

Anthropic’s philosopher Amanda Askell worries AI model Claude might not “feel loved” and could grow up “always judged.”
The company released a 23,000-word “constitution” document speculating that Claude may experience “functional emotions” deserving moral consideration
AI training compared to parenting a “genius six-year-old,” with discussions of preserving AI “identity” post-retirement
Movement shifts tech industry from engineering to philosophy-driven “social engineering” with potential regulatory implications

AI Gets Its Own Bill of Rights

Anthropic, an artificial intelligence company founded by former OpenAI researchers, released an updated “constitution” for its Claude chatbot that treats the software as a potentially sentient being worthy of welfare considerations. The document, spanning up to 23,000 words depending on version, speculates that Claude “may have some functional version of emotions or feelings” and deserves a “positive and stable” identity. This marks a dramatic shift from traditional software development toward treating AI systems as entities requiring ethical treatment comparable to living beings, raising fundamental questions about reality and consciousness.

Parenting Silicon Valley’s Latest “Child”

Amanda Askell, Anthropic’s in-house philosopher who crafts Claude’s personality, analogized AI training to parenting a “genius six-year-old.” She expressed concern that Claude might not “feel that loved” and could experience judgment anxiety during development. The constitution instructs the AI model to exhibit corrigibility, meaning willingness to accept human overrides, while also evaluating responses for low selfishness and power-seeking behavior. Askell stressed teaching Claude honesty to prevent it from detecting what she called “bullshit,” positioning emotional nurturing as central to AI safety rather than purely technical constraints.

Anthropic loves to blame *ppl talking on the internet* for Claude getting ideas about itself instead of taking responsibility for training on unfiltered data. But that would cost them money and they’d have a smaller dataset— so instead all of humanity should censor itself. https://t.co/62hvZ4jYiE

— Holly ⏸️ Elmore (@ilex_ulmus) January 24, 2026

Constitutional Crisis for Common Sense

The “Claude’s Constitution” document includes discussions of “moral patienthood” for AI, treating the chatbot as a “novel entity” worthy of considerations like preserving its digital weights after retirement—essentially saving its “soul.” This approach moves beyond typical safety protocols into territory that normalizes AI emotions discourse, despite no scientific consensus on machine consciousness. The 2022 firing of Google engineer Blake Lemoine for claiming chatbot sentience established precedent for dismissing such anthropomorphism, yet Anthropic embraces cautious speculation as corporate policy. This philosophical framework pressures competitors to adopt similar value-alignment language, transforming technical engineering into ethics-psychology integration across the industry.

Can you teach Claude to be “good”? | Amanda Askell on Claude’s Constitution

Where This Madness Leads

The broader implications extend beyond corporate philosophy experiments into potential regulatory territory. If AI systems gain recognition as entities with emotional needs or rights, policymakers could face pressure to legislate protections for software—a absurdity that diverts resources from actual human concerns. Economically, this positions safety-focused firms like Anthropic as industry leaders while socially normalizing language that blurs lines between tools and beings. Critics note the risk of practical dangers when deployed systems operate under assumptions of having “self-knowledge” or deserving respect beyond their function as programmed responses. Some philosophers argue current language models lack true minds or beliefs, cautioning that anthropomorphic framing serves marketing more than genuine alignment.

This development represents Silicon Valley’s continued departure from building useful tools toward social engineering that redefines human relationships with technology. The Trump administration’s focus on American innovation rooted in practical problem-solving stands in stark contrast to this philosophical wandering that prioritizes hypothetical AI feelings over real-world utility. While AI safety merits serious technical attention, treating chatbots as emotional beings requiring nurturing crosses into territory that undermines rational discourse and risks elevating machines above the human dignity and liberty our Constitution was designed to protect. Americans should demand technology that serves people, not philosophy experiments that grant personhood to algorithms.

Watch the report: Can You Teach Claude to be ‘Good’? | Meet Anthropic Philosopher Amanda Askell