Roadmap ⑨ · Available individually · €19 · web roadmap + A1 poster · Version 1.0 · As of 2026-08-19
Advanced Materials
From simulation to synthesis: AI finds candidates in weeks — manufacturing, qualification and regulation take years
1 Executive summary (excerpt)
For centuries materials research was a craft of trying, measuring and discarding. Since 2023 it has become a computational problem: with GNoME, Google DeepMind predicted 2.2 million crystal structures, 380,000 of them classified as stable; Berkeley’s A-Lab synthesised 41 new compounds in 17 days without a human hand at the furnace; machine-learned interatomic potentials compute millions of atoms with quantum-chemistry accuracy. In 2025 an MIT field study in an industrial lab measured the effect causally for the first time: 44 % more materials discovered, 39 % more patent filings — with falling job satisfaction among researchers, because judgement about candidates moves to the machine.
The bottleneck has thus moved, and to an uncomfortable place. Prediction is not synthesis: whether a material can be made — with which precursors, at what temperature, in which process window — is still predicted only unreliably by the models. And even a synthesised material classically needs ten to twenty years to reach series application, because qualification, standardisation and approval do not accelerate just because discovery does. The second finding: in this field regulation often decides more than physics. The PFAS restriction covers more than 10,000 substances and forces a material transition in semiconductor manufacturing, medical technology and energy that means years of qualification work; the Critical Raw Materials Act and China’s export controls on heavy rare earths make availability a location question; the ecodesign and construction products regulations make recyclability a selection criterion.
2 What you get
- Executive summary: what it is about, who it is relevant for, what the roadmap claims
- Roadmap panel: retrospective since 2020 and outlook today → 3 → 5 → 10 years in four layers — interactive, with rationale, dependency chains and confidence per entry
- ADL classification per reference industry (emerging · pacing · key · base)
- Maturity (TRL 1–9) per sub-technology with stated confidence and evidence
- Implications for action per horizon: watch · pilot · invest · ignore — with reasons
- Change journal and complete list of sources, methods used linked
- The A1 poster (594 × 841 mm, portrait) as vector PDF in English and German, plus PNG — with 50+ milestones in four layers, dependencies, key figures, TRL, ADL and action matrix
3 What a roadmap looks like
The freely accessible showcase roadmap Robotics & Humanoid Systems shows structure, depth and poster exactly as you get them here.
4 Questions about buying
- What exactly am I buying?
- Access to the web roadmap (permanent, tied to your e-mail address, sign-in by link without password) and the poster as PDF/PNG for download — as of the time of purchase (version and date are printed on page and poster). Later updates are not included.
- How does payment work?
- Via Stripe (card, Apple Pay, Google Pay, SEPA direct debit and others). You receive a receipt or invoice by e-mail. Price €19; no VAT is charged (German small-business rule, Section 19 UStG).
- Right of withdrawal?
- Digital content: by purchasing you request immediate provision and acknowledge that your right of withdrawal thereby expires — this is stated explicitly once more before payment. Details in the terms of purchase.
- For teams?
- Access is personal. For sharing within a company contact us: info@operation-zukunft.de.