AI & Robotics
Artificial Friend companion robot
A humanoid robot companion that observes a child closely, understands her emotions and devotes itself to her wellbeing and loneliness.
In shortArtificial Friend companion robot, from Klara and the Sun (Kazuo Ishiguro, 2021), is at technology readiness level 3 of 9 (experimental proof of concept) at its weakest link. Held back by Continual learning without catastrophic forgetting. Science status: underlying science well understood; an engineering problem. Estimated new funding to unlock: $134B–654B.
Klara, a solar-powered Artificial Friend chosen from a store, studies the sick teenager Josie's moods and is considered as a replacement for her if she dies.
Closest todayAI companion apps and 1X NEO-style home humanoids.
Dependency stack
Research TRL 1–3Development TRL 4–6Deployment TRL 7–9
Sort by
- Klara's core skill is reading human emotion accurately
✕3 blockers›
- Same expression maps to different emotions across contexts, cultures and individuals
- No ground-truth labels for internal emotional states
- Models learn demographic stereotypes from biased labels
TRL 4- Context and language understanding carry most emotional signal
✕2 blockers›
- Reliability and calibration over long horizons
- Continual learning after deployment
TRL 7- Training and inference compute
✕2 blockers›
- Grid power and interconnection queues
- Advanced-node chip and HBM memory supply
TRL 8
- A walking, safe household body
✕3 blockers›
- Cooperative safety certification for robots working in contact with people
- Battery life of 2–4 hours versus 8+ hour shifts
- Task generality beyond tote handling and pick-and-place
TRL 6- Hands and policies to handle arbitrary objects
✕2 blockers›
- Robot training data at scale
- Tactile sensing and robust hands
TRL 5- Fine force control and slip detection need touch
✕3 blockers›
- Elastomer skins wear out and need recalibration after thousands of contacts
- Wiring and bandwidth for thousands of taxels across a whole hand
- Few large tactile datasets to train manipulation policies
TRL 5 - Generalist perception and planningTRL 7
- Touch feedback for safe contact and fine manipulationTRL 5
- Full-shift untethered operation
✕2 blockers›
- Cycle life of lithium-metal anodes
- Thermal runaway safety
TRL 6
- Building a lasting relationship and memory of one family
- Weight updates on new data degrade prior tasks and safety alignment (spurious forgetting)
- No method to validate a model whose parameters keep changing after release
- Per-user weight updates multiply serving cost and memory
TRL 3- Online updates and replay at frontier scale need large training compute in the loopTRL 8
- Natural conversation and social reasoningTRL 7
- supportingScience of machine consciousnessThe novel treats Klara's inner life as real
✕3 blockers›
- Competing theories (global workspace, IIT, higher-order) make conflicting predictions
- Behavioral reports are trivially imitated by language models
- No ground truth for consciousness outside humans to calibrate tests
speculativeTRL 1- Need to know what brain mechanisms give rise to experience
✕1 blocker›
- Unknown sufficient level of biological detail
plausibleTRL 2- Wiring diagrams constrain and test circuit models
✕3 blockers›
- Human proofreading of AI segmentation dominates cost; 2025 methods imply tens of billions of dollars for one mouse brain
- Imaging throughput and storage; a human brain at EM resolution is of order 100+ exabytes of raw data
- Sectioning or expanding whole large brains without tissue loss or distortion
TRL 3- Ultrastructure must survive preservation before it can be imaged
✕3 blockers›
- Ischemic damage starts within minutes of circulatory arrest, before perfusion can begin
- Aldehyde fixation may destroy molecular state (phosphorylation, receptor conformation) that could carry memory
- No assay proves preserved tissue still contains the information an emulation would need
TRL 4 - Automated segmentation and tracing of exabyte-scale image volumesTRL 8
- Checking indicator properties requires reading internal model mechanisms, not behavior
✕1 blocker›
- Explanations cover only a fraction of model computation
TRL 3
Who could unlock it
- Across several requirements:
- Anthropic 2
Continual learning without catastrophic forgettingTRL 3bottleneck
- Ali BehrouzResearcher, Google Research (and Cornell University)First author of Titans (2025) and Nested Learning / Hope (NeurIPS 2025), the self-modifying architecture with a continuum memory system that reduced forgetting on continual-learning benchmarks
- Google ResearchCompanyPublished the Nested Learning paradigm (2025) and the follow-on "sleep" consolidation work (2026) that distils short-term into long-term memory
- Bing LiuResearcher, University of Illinois ChicagoLong-running lifelong-learning programme; his in-context continual learner with an external learner (InCA, 2025) is the standard non-parametric baseline
- TrajectoryStartupRaised a $15M seed in 2026 and then a Sequoia-led round to build a platform that keeps agentic models training on signals from live products
Affective computing (emotion inference)TRL 4
- Rosalind PicardResearcher, MIT Media Lab Affective Computing groupFounded the field (1995 book) and still leads wearable affect and stress-sensing research in 2025; co-founded Affectiva and Empatica
- Hume AIStartupShips an empathic voice interface (EVI 3 API, July 2025) that reads vocal expression and adapts its responses
- Smart Eye (Affectiva)CompanyAcquired Affectiva in 2021; sells facial-expression emotion sensing for in-car driver monitoring and market research, demoed at CES 2026
- EmpaticaCompanyPicard spinout whose wrist sensor became the first FDA-cleared AI smartwatch, for seizure detection, using physiological arousal signals
General-purpose humanoid robot platformTRL 6
- Agility RoboticsStartupDigit humanoid in paid warehouse deployments (GXO and others); raised a $400M round in 2025
- Figure AIStartupRaised $1B+ Series C at a $39B valuation (Sept 2025) to scale Figure humanoids with in-house Helix AI
- ApptronikStartupApollo humanoid; Series A grew to $935M+ with a $520M extension in Feb 2026
- Tesla OptimusCompanyTesla's in-house humanoid program, aiming at factory use and mass production (2025-2026)
General-purpose foundation modelsTRL 7
- OpenAIStartupGPT-5-series models (2025-2026) lead many agentic coding and reasoning benchmarks
- AnthropicStartupClaude Opus models (2025-2026) recorded some of the longest METR task time horizons
- Google DeepMindCompanyGemini 3 (Dec 2025) and Gemini 3.1 Pro (2026) are frontier multimodal reasoning models
- DeepSeekStartupChinese lab whose open-weight R1 reasoning model (Jan 2025) matched frontier results at reported low training cost
Beyond engineering
- Child attachment and emotional-dependence risks of AI companions