Dexterity Unveils Foresight: World Model for Physical AI Truck Loading
Physical AI company Dexterity Inc. has introduced Foresight, a physics-consistent world model and 4D packing agent designed to power autonomous truck loading and other industrial robotics applications, announced in a press release.
Foresight enables robots to perceive, reason, and act in real time using a transactable simulation of the physical environment. The system supports Dexterity’s dual-armed robot, Mech, which performs 4D box packing—considering three spatial dimensions and time—to determine optimal placements for packages inside trucks. Each decision takes under 400 milliseconds and optimizes density, stability, reachability, and dual-arm coordination.
The Foresight architecture connects perception, decision, and motion agents operating asynchronously and is designed to be interpretable and safety-focused. It has already been deployed across six applications and trained on more than 100 million autonomous production actions.
Dexterity also announced the Foresight API Challenge, inviting student teams to build packing agents and compete for up to $50,000 in prizes. The competition opens in March and includes a public leaderboard and a browser-based truck loading game available on dexterity.ai.
We hope you enjoyed this article
Consider subscribing to one of our newsletters like Robotics Brief, Industrial AI Weekly or Daily AI Brief.
Also, consider following us on social media:
More from Robotics
Nov 19 Wallenberg Investments takes a stake in Swedish simulation company Algoryx Nov 19 Helsing scales back UK underwater drone plans after missing out on £90 million Royal Navy deal Nov 19 MegaRobo raises nearly $200 million to let AI run real lab experiments Nov 18 Rise Automation raises CAD 2.4 million to bring robots to homebuilding factories Nov 18 Botsync raises $11 million to add AI to its factory and warehouse robotsRobotics Brief
Weekly coverage of AI-driven robotics advances in industrial automation, autonomous vehicles, and robotic systems.
Whitepaper
Tensordyne Napier: What If One Rack Could Do the Work of Nine?
Tensordyne
This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.
Read moreYou may also like
TÜV SÜD and Accenture Present Physical AI Safety Suite
Eureka Robotics launches StereoPick to let factory teams set up 3D bin picking without code
Dallas AI startup 4Minds raises $15 million at a $500 million valuation
Google DeepMind's Gemini Robotics 2.5 Hits 96% Success on Mercedes-Benz Lines
Serve Robotics Launches Physical AI Services for Robot Makers
Daily AI Brief: the AI news that matters, in your inbox.