Autonomous Driving

Today's Stories

1. 1

   Tech stocks rallied on AI infrastructure demand and U.S. chip strategy, as Apple-Intel partnership and major M&A deals signaled growing focus on domestic semiconductor capabilities.

   Intel gained nearly 5% after President Trump announced Apple would collaborate with the company on U.S.-based chip development and manufacturing. Separately, Nebius completed its acquisition of AI optimization specialist Eigen AI and posted a 10% weekly gain, while RUM Group finished its purchase of Germany-based Northern Data and launched its Quake AI platform. Ondas announced its sixth acquisition of 2026, agreeing to buy Cyberhawk for about $125 million（約200億円） to expand its drone inspections and critical infrastructure monitoring capabilities. Investors are watching companies that control hardware for AI computing and domestic chip production as governments and corporations prioritize securing their own semiconductor capabilities. The flurry of deals and partnerships reflects a strategic shift toward building AI infrastructure at scale and reducing reliance on overseas manufacturing, which appears to be reshaping where investors see long-term demand.

   While most semiconductor and AI infrastructure stocks declined during the week—UBER, TSLA, and ASTS fell between 1% and 7%—retail investor interest surged on some names (RUM message volume jumped 86%, MRVL message volume up 12%). Geopolitical tensions between the U.S. and Iran also lingered, though negotiators reported meaningful progress during talks in Switzerland on Monday, with hopes to finalize a broader agreement within the next two months.

2. 2

   Stratechery is a subscription-based news and analysis service offering daily business technology coverage, podcasts, and interviews — not a news story requiring summary.

   The article body provided is a subscription landing page describing Stratechery's offerings (daily updates, interviews, and podcasts) and FAQ, not a news article about a business event or announcement. There is no news event to analyze. The body contains only service descriptions, pricing ($15/month or $150/year), and subscription management instructions — it does not report a change, announcement, or development in business or technology.

   No forward-looking claim or date is present in the body. To produce a meaningful summary, a news article body (not a service landing page) is required.

3. 3

   DeepSeek releases R1, a reasoning model that matches top competitors while using significantly fewer computational resources, reshaping how AI labs prioritize efficiency over raw scale.

   DeepSeek released R1, a reasoning model (AI trained to work through problems step-by-step) that achieves performance comparable to OpenAI's o1 while requiring substantially less computational power to run. The model is available open-source, meaning anyone can download and use it. The release challenges the long-held assumption in AI development that bigger always means better. DeepSeek's efficiency suggests that smart engineering and training methods can compete with massive-scale approaches, potentially shifting where companies choose to invest resources and how they evaluate their competitive advantages.

   R1 is immediately accessible to developers and researchers as an open-source model, which means rapid independent testing and potential widespread adoption could reveal whether efficiency gains hold up across diverse real-world applications, or whether certain tasks still require the scale of closed commercial models.

4. 4

   AWS is showcasing Self-Driving Lab technology at its 2026 Healthcare & Life Sciences summit, presenting a vision of how physical AI (systems that combine computation with robotic automation) can reshape drug discovery.

   AWS is featuring Self-Driving Lab—an approach that integrates artificial intelligence with automated laboratory equipment—at AWS Summit 2026's Healthcare & Life Sciences track. The company is demonstrating how this technology applies to drug discovery workflows. Drug discovery traditionally relies on manual experimentation and human decision-making, which is time-consuming and expensive. By combining AI with robotic lab systems, Self-Driving Lab aims to automate hypothesis testing and experimental cycles, potentially accelerating the pace at which new drugs can be identified and validated.

   This is part of AWS's broader push to position itself in the life sciences and healthcare sector. The specific capabilities, timeline for broader availability, and which pharmaceutical or biotech partners may adopt this technology remain key indicators of how far physical AI can move from demonstration to production use in drug development.

5. 5

   Claude Code's human-approval feature is frustrating users because it requires repeated decisions on similar tasks, lacks context to make informed choices, and offers no way to delegate decisions to specialists.

   A user reported that Claude Code's approval workflow forces them to repeatedly say yes or no to similar requests across sessions—for example, being asked to use pip install despite having previously rejected it in favor of uv sync in the same project. The tool also asks for approval on complex changes (like database migrations that drop and recreate columns) without letting the user ask clarifying questions beforehand, and does not show the full impact of different options until after a choice is made. The repetition and lack of transparency mean users end up approving actions they don't fully understand, turning the safety mechanism into a bottleneck rather than a meaningful safeguard. When Claude Code asks for approval on a risky database operation, the user cannot ask "what does this actually touch?" or compare the consequences of different approaches before committing—they either approve blind or reject and restart, wasting time.

   The user notes there is no way to forward decisions to specialists (such as routing a database migration to a DBA); the only options are approve, reject, or manually contact colleagues outside the tool. This gap suggests the approval system may not scale to team workflows where decisions should route to the right person.

6. 6

   Waabi CEO Raquel Urtasun, who has raised over $1 billion（約1600億円） for her autonomous trucking company, argues that Gen Z workers with adaptability and curiosity—not decades of industry experience—are the talent driving AI transformation.

   Raquel Urtasun, cofounder and CEO of autonomous trucking unicorn Waabi, told Fortune she prioritizes hiring young workers who are versatile and eager to learn over candidates with 20 years of industry experience. Her company, launched in 2021 and funded through a recent Series C round co-led by Khosla Ventures, is already testing autonomous trucks with Volvo. Urtasun, who worked with Geoffrey Hinton in academia and headed Uber's self-driving division before founding Waabi, believes the most valuable skill is not mastery of a single specialty but the ability to learn and adapt as technology evolves. She views the current AI era as an opportunity rather than a threat to workers entering the job market.

   Waabi's autonomous trucks are already in road testing with Volvo, demonstrating progress toward a driverless future. McKinsey's recent survey found expectations for fully autonomous long-haul trucking have continued to slip, with timelines now stretching closer to the end of the decade—the competitive and regulatory landscape that will determine how quickly Waabi's technology reaches deployment at scale.