MSDINTERNATIONAL← Home Discover MSD ⌄From AI and robotics to islands, yachts, specialist partners and new business models. Choose your entry point – and discover what lies behind it.
AI, robotics & futureFuture · Die Firma aus Systemen10,000 AI agents as a scaling target, networked tasks and an interactive estimator.
↗
AI, robotics & futureRobotics · 10 platformsManufacturers, capabilities, interfaces and integration tasks.
↗
MSD Direct & MarktplatzMSD Direct · Das PortalPeople, opportunities, way of life.
↗
MSD WORLDSiteWarden · Robotics LabGripping, sensor technology and physical AI – planned experimental programs.
↗Die Übersicht wird geladen.
MSD / Grundlage / RESEARCH EDITION
The Transformer paper presents an attention-based architecture for sequence processing. It is a methodological building block of later language and multimodal models.
Jetzt eintauchen ↓RECHERCHESTAND 12.11.2026 · FAKTEN + EINORDNUNG + SZENARIENKI & Agenten
External research · MSD classification
Keine Behauptung eines eigenen Versuchs.
01 / EINORDNUNG
An architecture alone does not guarantee reliable facts or safe actions in a company.
Design model capabilities, available tools, and allowed business actions as three separate layers.
This derivation is an editorial recommendation from MSD. It is not a further result of the study cited.
Vaswani et al. · Attention Is All You Need · Publikations-/Bezugsjahr 2017. Recherchiert am 12.11.2026.
Read primary source ↗External research, not an MSD customer reference. We will only publish our own experiments with their setup, date, measurements and documented failed attempts.
02 / LINKS
In a randomised trial with 16 experienced developers and 246 tasks, early-2025 AI tools increased completion time by 19%. Participants perceived their benefit much more favourably.
METR · Randomised trialMETR describes a follow-up experiment in February 2026. Evidence of faster work is difficult to interpret due to selection of participants and tasks as well as problems with time measurement.
METR · Methodenupdate 24.02.2026ReAct examines how language models can connect intermediate steps with actions in an environment. Tools provide new observations for further steps.
Yao et al. · ReAct