250-fold profit surge reveals the AI memory shortage squeeze—44% DRAM, 53% NAND price hikes in Q2 2026.
Category: robotics/AI – Page 38
NASA’s Starling Mission Opens New Frontiers in Space Navigation
NASA’s Starling mission has marked another milestone in spacecraft autonomy by using a new system that determines a satellite’s position in orbit by referencing other objects in space, instead of relying on a navigational network.
The FALCON (Fast Autonomous Lost-in-space Catalog-based Optical Navigation) technology demonstration is a step toward spacecraft being able to operate more independently. As NASA prepares for more missions beyond Earth’s orbit, technologies like FALCON can support lunar satellite swarms, distributed science missions, and human exploration.
Traditional satellite navigation depends on GPS signals, but those can be unreliable or unavailable in lunar or deep space environments. The FALCON payload is a joint flight experiment by NASA and EraDrive, a startup spun out from Stanford University. It combines EraDrive’s Era-Core flight software and embedded algorithms with Starling’s cameras and an onboard catalog of known satellites to support GPS-independent navigation and space situational awareness.
Humanoid Robot Production Numbers: The 50,000-Unit Claim Nobody Can Verify
Widely repeated humanoid robot production numbers, Tesla past 50,000 Optimus units, Figure AI past 10,000 deployments, don’t come from the companies themselves. Tesla’s own 2025 target was 5,000 units for internal use; actual output was a few hundred. The gap between claimed and audited numbers is now the story investors and buyers should be pricing in.
Humanoid robot production numbers have become the least reliable statistic in industrial AI. Search for how many humanoid robots are actually working today and you’ll find confident figures everywhere: Tesla past 50,000 cumulative Optimus units, Figure AI past 10,000 deployments across partner warehouses. None of those figures come from the companies they describe, according to a July 2026 investigation by Technology.org. Tesla has never published an audited Optimus production count.
The clearest example sits in Tesla’s own record. The company targeted 5,000 Optimus units for internal factory use in 2025. Actual output, per mid-2025 reporting from The Information cited by Technology.org, was a few hundred units, under ten percent of the goal. Contributing factors included China’s April 2025 export restrictions on rare earth magnets and a longer-than-planned Gen 3 design finalization, particularly on the dexterous hand system.
How the Department of War Is Rebuilding Its Innovation Edge
Chuck Brooks is the president of Brooks Consulting International and one of Executive Mosaic’s GovCon Experts.
The United States is about to enter a new age of technical competitiveness where the rate of innovation and invention could have just as much of an impact on national security as the technology itself. Emerging technologies such as directed energy, biotechnology, autonomous systems, robotics, advanced manufacturing, artificial intelligence and quantum computing are developing concurrently and rapidly converging.
This convergence is posing a fundamental dilemma for the Department of War: Can the government organize itself to find, develop, buy and field new technology at the same rate as the business sector and America’s adversaries? A major reorganization of the Department’s research, technology and innovation ecosystem is starting to reveal the solution.
AI Cybersecurity Access Tiers: Who Gets The Best AI?
OpenAI’s new Daybreak Red tier gives approved defenders a model that completes 95% of advanced exploit-development requests, versus roughly 2% for the public version of the same base model. That gap is now the real story: AI cybersecurity access tiers decide who gets frontier defensive power and who doesn’t, and access runs through a partner list, not a price tag.
AI cybersecurity access tiers stopped being a theoretical debate this month. OpenAI expanded its Daybreak program into two levels, Blue and Red, and released GPT-5.6-Cyber, a specialized model built specifically for vulnerability research and exploit-chain development, according to SecurityBrief’s coverage of the launch. In an internal OpenAI evaluation, the new model completed 95.0% of advanced cyber requests covering authentication bypass, privilege escalation, and exploit-chain development, compared with 1.5% for the general-release model and 2.0% for that same model through the safeguarded Daybreak Blue tier.
GPT-5.6-Cyber is only available through Daybreak Red, gated behind identity verification, monitoring, legal attestations, and approved-use restrictions, per Cyberpress’s reporting. Under OpenAI’s own Preparedness Framework, the model was rated “High” for cybersecurity capability, one step below the “Critical” threshold that recently triggered an internal suspension of a different unreleased model, Astra, on August 7. These AI cybersecurity access tiers exist because the underlying capability is real: a general-purpose model built to refuse exploit-writing requests is far less useful to a security team validating a patch than one built to complete them.