Announcing “Computational Functionalism Debate” (soliciting paid feedback): Test your intuitions about consciousness

We’re excited to announce “Computational Functionalism Debate” (cf-debate.com), a website containing the largest available assembly of arguments in support of and challenging digital consciousness (42 arguments and counting).

The goal is to offer a centralised, easy to navigate repository of the cruxes commonly mentioned in debates about computational functionalism in a digital context. Computational functionalism is a philosophical position that unlocks key theories of consciousness used to assess AI model welfare today. It is plausible but uncertain, defined at a high level as: "performing computations of the right kind is necessary and sufficient for [phenomenal] consciousness." (More detail here on definitions, which philosophical positions matter and which can be ignored for AI consciousness.)

To improve the website content, we are launching a reward program: Major additions or edits prompted by reader input will be eligible for $100 USD, paid directly to you or to a charity of your choice (smaller changes might attract smaller rewards).

This is a project of the Co-Sentience Initiative, a soon-to-be-launched initiative with the goal of diversifying the ecosystem of consciousness research to prepare for a possible future shared with artificial minds. In 2026, we plan to launch new research building on this site, e.g. on which arguments are found most convincing by different audiences, which arguments are least well-known, and what’s typically happening when people change their mind about consciousness (and how often this actually happens). You can sign up for updates via the website.

Test your intuitions

You can test your intuitions about computational functionalism (CF) with our quiz.

Our early research suggests that most people have split intuitions. Even if they support CF in principle and most arguments against CF feel reasonable to them, there will be a few arguments where they lean the other way - and vice versa. These contrary-intuitions are particularly valuable for stress-testing or refining your views.

Interrogating your intuitions - looking for conflicts and identifying consistent narratives that reconcile them - is a fun way of making progress. Exact certainty might be out of reach in philosophy, but some arguments are materially better than others and revised views tend to be more convincing than prior views. The quiz read-out will suggest a few arguments personalised to where your intuitions point in different directions or where you are particularly uncertain - a starting point for exploration.

Some of the stronger arguments in favour of CF include the below, but check out the website for the fuller exposition & some common counter-arguments:

• Church-Turing thesis extension: All functions can be emulated/represented on any big-enough computer, so anything that matters about consciousness can be done computationally.
• Multiple realisability: Very different animals appear to experience pain, so experience must be defined by how systems are organised not what they are made out of.
• Fading qualia: If I keep replacing your biological neurons with functional equivalents in silicon one-by-one, your behaviour would stay the same and you would stay conscious.
• AI success argument: AI is proving it can master the whole range of human capabilities; consciousness isn't any different.

Likewise, here are some arguments against CF - but again, there’s a live debate to be had about each one. We’re cataloguing responses and responses-to-responses: keep sharing them with us.

• Physics violations: The fundamental equations of physics depend only on the current state of the universe, so consciousness cannot be based on functions defined by multiple past states.
• Phenomenal binding: Digital computation reduces entirely to simple, separable 0/1 operations; there's no space for a complex, unified, causally-relevant macro-experience like humans have.
• Individuation problem: No single step of an algorithm has visibility of the whole algorithm, so the 'whole process' only exists in the eyes of the program user, i.e. us.
• Pen & paper: CF implies that any conscious experience can be generated by doing the calculation by hand on paper, even if it takes thousands of years. Such experiences also have no causal relevance over themselves or anything else.

Inclusion of an argument does not mean it provides fatal evidence against CF or compelling evidence for it. Arguments vary in strength and the total number of arguments does not reflect the overall strength of a position. The key point is that a robust view requires a coherent position that responds to all arguments challenging it, not just the easy ones. Some of these debates have been raging for centuries, so it's worth being open to the possibility that there are counter-arguments available to your responses, but that doesn't mean you couldn't overcome them…

Reward program

We want our collection of arguments to be as complete and usable as possible. To that end, readers are invited to submit suggestions for edits, additions, or improvements.

Major additions or edits prompted by reader input will be eligible for $100 USD, paid directly to you or to a charity of your choice (smaller changes might attract smaller rewards).

What kind of feedback are we looking for?

• New arguments that should be included (we’re at 42 and counting…)
• Corrections to existing argument descriptions
• Additional responses or counter-arguments
• Better categorisation suggestions
• Improvements to clarity or neutrality
• Technical issues or usability problems

The current project lead is Chris Percy PhD, supported by a small team of experts in analytical philosophy, computer science, and meta-ethics. You can contact Chris at [email protected] or on X via @chris_percy — anonymous feedback can also be submitted online. We’re also happy to answer questions via the comments on this post.