Observed Signal · Aug 6, 2026 · Analysis · Source: t3n · Impact: 3/5 · Sentiment: Neutral
How AI Is Learning to Read Emotions from Faces
A feature examines how artificial intelligence systems attempt to recognize human emotions from facial expressions and speech. Björn Schuller, a researcher at TU Munich and Imperial College London, describes using an Arousal-Valence-Dominance model and the six basic emotions as classification points. The article notes criticism of the scientific basis for current emotion-recognition systems, cites market estimates (about $40 billion revenue last year from emotion recognition with projected strong growth to 2034), and discusses multimodal foundation models, studies showing improved performance over humans in some tests, and young people sometimes preferring chatbots for emotional disclosure.
Emotion-recognition AI intersects privacy, measurement validity, and potential new targeting/insights use cases for marketing; the market size estimate (~$40B) and projected growth make it materially relevant to AdTech and MarTech players.
Track MIT Technology Review Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- Researcher Björn Schuller uses a three-dimension Arousal-Valence-Dominance model to map emotions and the six basic emotions (anger, disgust, sadness, surprise, fear, joy).
- Emotion-recognition systems are trained on human-labelled datasets and estimate perceived emotions rather than measuring objective internal states.
- Fortune Business Insights estimated emotion-recognition startups and companies generated about $40 billion USD worldwide last year (approx. €35 billion) and projected the sum to more than triple by 2034.
- A University of Bern study found large multimodal models performed significantly better than humans in tests of emotional intelligence.
- A March study by DAK Gesundheit and the Universitätsklinikum Eppendorf found about one in ten adolescents prefer confiding in a chatbot rather than a friend.
Connected Companies & Entities
5 Entities mapped“Dieser Text erscheint in der Ausgabe 6/2026 von MIT Technology Review....”
“Title: Wut, Trauer, Freude: Wie KI unsere Emotionen im Gesicht erkennen soll (page hosted on t3n.de)...”
“Im letzten Jahr setzten Startups und Unternehmen wie Amazon, Microsoft oder IBM laut Fortune Business Insights weltweit gut 40 Milliarden US...”
“Im letzten Jahr setzten Startups und Unternehmen wie Amazon, Microsoft oder IBM laut Fortune Business Insights weltweit gut 40 Milliarden US...”
“Im letzten Jahr setzten Startups und Unternehmen wie Amazon, Microsoft oder IBM laut Fortune Business Insights weltweit gut 40 Milliarden US...”
Ontology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
Startups Aim to Make AI More Human
A wave of early-stage startups is building AI agents designed to simulate human emotions, intentions and behavior. Simile raised a $100 million round led by Index partner Shardul Shah after training models on recorded conversations and behavioral-science data to predict emotional reactions. New ventures named in the piece include Prior Computers (founded by researchers from MIT and Harvard), People Make Things (stealth, focused on intent prediction), Aaru (valued near $1 billion in a recent round led by Redpoint), Humans& (founders from Google, Anthropic and xAI with $480 million of backing), Expected Parrot (an open-source interview dataset), and Constellation Systems (seeded to build a foundation model of “human state”). The coverage highlights investor momentum and varied technical approaches to modeling human-like responses for applications such as market research, customer simulation and collaboration tools.
Interhuman AI teaches AI to read non-verbal human signals
Copenhagen-based Interhuman AI has launched Inter-2, a model that detects 12 social signals such as hesitation, uncertainty, and agreement in real-time from text, audio, and video. The startup aims to introduce Artificial Social Intelligence, enabling AI to understand non-verbal cues. Co-founded by CEO Paula Petcu, COO Frederik Sally, and CRO Line Clemmensen, the company is building its own models in Europe. Inter-2 delivers up to four times faster inference with improved benchmark performance. The technology is targeted at applications in corporate training, digital health, sales coaching, and market research. Interhuman emphasizes transparency and traceability of model outputs, adhering to GDPR and EU AI Act. The company is expanding into voice-only interactions and will announce two additional models in October.
Study: Emotions Boost Advertising Effectiveness
A Screenforce and Eye square study (reported by W&V) finds that emotional advertising significantly improves memory, liking and purchase intent, but must be carefully dosed: too little emotion bores viewers, too much overwhelms them. The research filmed participants’ faces during media consumption and applied software plus an AI based on Semantic Space Theory to detect fine‑grained facial micro‑movements and emotional states. Results show positive emotional cues (humor, familiar music, strong storytelling, likable protagonists) lift recall and buying intention, while negative reactions (provocation, irritation, shock) reduce attention — on average about five seconds less — and lower purchase intent. The authors recommend integrating emotional measurement early in creative development because neglecting emotional impact can raise overall media costs.
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.