Emotion intelligence · Private beta

Know what your images
do to the brain.

Amphora gives generative AI a biological basis for feeling. Our fMRI prediction engine maps any visual input to predicted brain activation — turning emotional outcome from a guess into a specification.

You're in. We'll be in touch before your access opens.
Something went wrong — please try again or email us at hello@amphora.ai
Developed by students at
Harvard University
Stanford University

From the lab · May 2026

We fine-tuned a language model using real brain signals. It worked.

Using TRIBE v2 — Meta AI's fMRI encoder — as the reward signal, we steered a 3B parameter language model toward higher predicted cortical activation in Broca's area over 200 RL training steps. The entire language network responded. Here's the data.

Cortical surface maps — base, LoRA, difference Cortical surface — base · LoRA · Δ
+150%
Broca's area reward gain over 200 training steps
×4.6
Global mean cortical BOLD increase (0.023 → 0.104)
20
Cortical regions measured — 19 of 20 showed positive gains
Training trajectory
Reward · loss · KL over 200 steps
ROI bar chart
20 cortical ROI activations — base vs LoRA
All ROI timeseries
BOLD timeseries — all 20 regions

The model learned to write differently. Without being told how, it shifted from scene-setting prose to dialogue-driven narrative — exactly the style that predicts higher BOLD in Broca's, Wernicke's, and STS regions.

Read the full write-up →

Progress Notes · Members only

What we're learning, in writing.

Enter your email to read →
May 26, 2026 Research

We Fine-Tuned a Language Model Using Brain Signals. Here's What Happened.

We used TRIBE v2 as the reward signal in an RL training loop. After 200 steps, the model had learned to produce text that drives 150% higher predicted cortical activation in Broca's area.

May 23, 2026 Essay

Emotion as a Specification, Not a Judgment

The generative AI industry has spent enormous effort making images more realistic. It has not given AI any mechanism to understand what an image does to the person looking at it.

May 18, 2026 Progress

The fMRI Prediction Engine: Phase 1 Complete

The core of what we're building — a lightweight model that takes any visual input and outputs predicted brain activation across emotional and perceptual regions — is working.

May 10, 2026 Vision

Why We're Building Amphora

The generative AI industry has solved for realism and style. It has not solved for feeling. That gap is where Amphora lives.

01

Neural Prediction Engine

Submit any visual asset — a photo, AI render, ad creative — and receive a map of predicted brain activation across the regions that drive emotional and perceptual response. Not vague labels. Precise neuroscience.

02

Emotion Guidance Loop

Integrate our API into any generative AI pipeline. Set a target emotional profile — aspiration, intimacy, awe, tension — and let the system iterate until the output scores for the intended feeling. Emotional coherence becomes a specification, not an afterthought.

03

Scientific Creative Signal

Know whether your image activates the precuneus for aspiration, the amygdala for threat and awe, or the fusiform cortex for intimacy and recognition — before it goes live. A precise, scientifically grounded signal for designers, AI systems, and brand strategists.