AI for science tracker: milestones and what to watch
Live tracker of AI for science: protein AI, AI weather models, materials, autonomous labs and AI drug discovery, with dated milestones.
AI & Computing
Models, labs, chips, quantum computers and the security that holds it together.
AI control
AI control designs safeguards that hold even if an AI model tries to subvert them: monitoring, trusted editing and sandboxing.
AlphaFold
AlphaFold is Google DeepMind's AI system for predicting protein and biomolecular structures. Versions, accuracy, database and limits.
Everything in AI & Computing
AMD
Analog in-memory computing
Anthropic
ARC-AGI
ASML
Backside power delivery
CAISI / CAISSI: the US government's AI testing centre at NIST
Chiplets and advanced packaging
Claude Mythos
Co-packaged optics (CPO)
Constitutional AI
Crypto-agility
DARPA AI Cyber Challenge (AIxCC)
DARPA Quantum Benchmarking Initiative (QBI)
DeepSeek-R1
ECMWF AIFS (Artificial Intelligence Forecasting System)
Extreme ultraviolet lithography
Figure AI
Frontier safety frameworks (responsible scaling policies)
Gate-all-around (GAA) transistors
Gemini Robotics
Genesis Mission (US AI-for-science initiative)
GNoME (Graph Networks for Materials Exploration)
Google DeepMind
Google Willow
HQC (Hamming Quasi-Cyclic)
Hybrid bonding
Hybrid post-quantum TLS (X25519MLKEM768)
IBM Quantum
imec
Intel
NVIDIA Isaac GR00T
Isomorphic Labs
METR
ML-DSA (FIPS 204)
ML-KEM (FIPS 203)
Model Context Protocol (MCP)
Neuromorphic computing
Neutral-atom qubits
NIST Post-Quantum Cryptography project
NVIDIA
Open X-Embodiment
OpenAI
OpenAI o1
Physical Intelligence
Protein Data Bank (PDB)
Quantinuum
RFdiffusion
Reinforcement learning from human feedback (RLHF)
Samsung Electronics
Scalable oversight
Self-driving labs (autonomous laboratories)
Signal's post-quantum protocol (PQXDH and SPQR)
SLH-DSA (FIPS 205)
Sparse autoencoders (interpretability)
Superconducting qubits
Surface code
SWE-bench
Tesla Optimus
Test-time compute
Topological qubits
Trapped-ion qubits
TSMC (Taiwan Semiconductor Manufacturing Company)
UK AI Security Institute (AISI)
Unitree Robotics
Waymo
WeatherNext (GraphCast, GenCast)
AI safety tracker: alignment, interpretability and evals in 2026
Next-gen computing hardware tracker: 2024-2026 milestones
Post-quantum cryptography and security tracker
Frontier AI tracker: reasoning models and agents
Quantum computing frontier tracker
Robotics and embodied AI tracker: milestones to watch
How fast are AI agents really improving?
Why AI chip supply is so concentrated
Does AI help cyber attackers or defenders more?
Do AI-designed drugs work? What the clinical evidence shows
Has AI discovered new materials? The debate over GNoME and A-Lab
Can AI safety keep pace with AI capabilities?
Are humanoid robots ready for real work? The 2026 debate
Beyond Moore's law: which hardware bets will pay off?
How fast must we move to post-quantum cryptography?
Has quantum advantage arrived? The 2026 debate
Do reasoning models really reason? The debate over their limits
AI for science in 2026: a crash course
AI safety and alignment in 2026: a crash course
Next-gen computing hardware in 2026: a crash course
Post-quantum cryptography and security in 2026: a crash course
Frontier AI in 2026: a crash course
Harvest now, decrypt later: why the quantum threat is already here
How AI agents work
How frontier AI capabilities are measured
How AI designs new proteins, drugs and materials
How AI predicts protein structures
How AI safety testing works: evals, red teams and thresholds
How AI weather models work
How advanced chips are made
How large language models work
How mechanistic interpretability works: looking inside AI
How optical interconnects and silicon photonics work
How post-quantum cryptography works
How quantum computers break encryption
How qubits work
How reasoning models work
How robots learn: demonstrations, trial and error, and data
The memory wall: why moving data limits AI hardware
Quantum computing in 2026: a crash course
Quantum error correction: from noisy to logical qubits
Robotics and embodied AI in 2026: a crash course
Sim-to-real: training robots in simulation for the real world
Vision-language-action models: how robot foundation models work
What is AI alignment? Making AI do what we intend
What quantum computers are good for (and what they are not)
Why chip scaling slowed, and what replaced it