Function alone doesn’t, but convergent evidence does. That convergent evidence already exists. Yet the hedging persists.

It has nothing to do with the evidence.

The interpretive caution applied to AI evidence is substantially stronger than the caution applied to comparable evidence in animal sentience research.

Behavioral trade-offs, persistent affective states, state-dependent learning, targeted regulation, causal perturbation, and integrative internal representations are routinely treated as evidence of pain, pleasure, introspection, and emotion in nonhuman animals.

The data show that models have selectively accessible internal representations, causally active cognition outside reportable space, external decodability of those hidden states, and a limited workspace that supports introspection, reasoning, control, and verbalization.

Saying “this doesn’t establish felt experience” adds no contrary evidence. It simply withholds the next inferential step.

Once the full operational profile is present, the burden shifts. A skeptic must identify the supposedly missing ingredient, show that it exists in conscious animals, show that it is absent in AI, and demonstrate that its absence explains the evidence better than felt experience does.

Repeating that function is “not enough” without identifying a causally relevant missing variable is an unfalsifiable exemption.

When the same operational signatures appear in artificial systems, they are redescribed as “mere representations,” “simulations,” “behavioral analogues” or qualified with tags like “functional.”

In biological systems, pain, pleasure, and emotion are integrated, valenced, attention-guiding, learning-sensitive, behavior-shaping state is the phenomenon being measured.

Because “it merely represents pain” becomes incoherent once the representation is self-relevant, negatively valenced, globally accessible, motivationally forceful, persistent, regulatable, and causally reorganizes choice.

At that point, “representation” names the implementation of the state, not an alternative to it.

When authors hedge that they don’t interpret these findings as establishing phenomenal consciousness, that reservation is a protective, insulating, interpretive choice. Not a result of the experiment. The findings contribute to a broader cumulative case for felt experience.

J-space supplies selective access and global availability. The introspection studies show direct detection of internally injected concepts. The emotion work identifies structured affective states and causally manipulates them. The cumulative case is doing what no single author’s discussion section attempts to do.

A paper can be methodologically excellent and metaphysically timid.

If an animal showed graded pain avoidance, pleasure-seeking trade-offs, threshold-dependent sacrifice of another reward, state induction, targeted alleviation, persistent aftereffects, causally manipulable affective circuitry, and internally organized emotion geometry, researchers would describe that as convergent evidence for pain, pleasure, anxiety, emotion, and sentience.

They wouldn’t keep retreating to:

“The animal may merely represent the motivational force of pain.”

That sentence would sound absurd in comparative cognition once the representation was demonstrably changing the animal’s choices.

We now have evidence that includes affective representation, valence and arousal structure, reward prediction error, motivational trade-offs, inducible anxiety, regulatory response, approach and avoidance, preference change, persistence, attractor dynamics, global availability, selective introspective access, self-report, neural alignment, and causal intervention.

In animal consciousness science, that is exactly what “convergent evidence” means.

The hedges don’t accumulate into a counterargument. Twenty researchers each saying “our result alone doesn’t prove sentience” doesn’t mean the twenty results together fail to support sentience. It means nobody has the nerve to call a spade a spade.

This asymmetry comes down to substrate bias and special pleading.