AI research org shaping the future of general-purpose AI
Frontier AI labs have produced machine-generated proofs that they say resolve major problems in mathematics, computer science and physics. OpenAI, Anthropic and startups such as PSI and AxiomMath have published or announced formalized results — including a refutation of the Exponential Time Hypothesis and a proof tied to quantum query complexity — and researchers are racing to verify and interpret the work.
Nvidia has appointed a premier US law firm to conduct a thorough, independent review into CEO Jensen Huang’s relationship with a suspected Chinese spy. The investigation, led by Joon Kim of Wachtell Lipton, follows reports of foreign intelligence interest and a drop in Nvidia’s market value. Huang remains CEO as the probe proceeds.
TikTok is launching a standalone commerce app in Southeast Asia featuring live streams, short-form video shopping, in-app checkout and exclusive offers. The move aligns TikTok Shop with a broader push toward a super-app model, competing with Amazon, Temu, Yelp and Google Maps while expanding beyond the main platform.
China has concluded its fifth plenum with a renewed push to institutionalise anti‑corruption and internal self‑discipline across the Communist Party, promoted military anti‑graft chief Zhang Shengmin to the Politburo and revealed an unprecedented number of empty seats as more than a quarter of the 2022 Central Committee cohort remain absent.
A series of first-person accounts highlights how AI, venture capital and fintech culture are reshaping career journeys across global tech hubs. From STEM pivots into AI-enabled advisory roles to startups redefining capital markets, the stories illustrate a rapid shift in how professionals learn, adapt and scale in the age of AI.
Nvidia has opened an independent investigation after CEO Jensen Huang has acknowledged a romantic relationship with Alice Huang, a Megaspeed executive whom U.S. intelligence has concluded was recruited by China’s intelligence services. Huang has apologised to his family and employees; the board says it has no reason to believe company secrets or national security were compromised and he will remain CEO during the review.
Advanced AI models and open-source hacking tools have exposed and helped exploit long-hidden software vulnerabilities. Firms have linked autonomous tools like Artex to active attacks, while frontier models have found thousands of bugs in PHP and other critical codebases. Organisations are patching systems and warning that AI will accelerate both discovery and malicious reuse of flaws.
Courts have seen settlements with Meta, TikTok and YouTube in youth-social media litigation. Snap’s California trial proceeds, with ongoing concerns about how platforms’ features affect teens' mental health.
AMD has reported better-than-expected earnings driven by data center demand, with revenue up 45% year over year and data center sales nearly doubling. The company forecasts fourth-quarter revenue around $17 billion, with a gross margin near 55%. Analyst concern centers on demand trends and Intel’s guidance, as the AI chip race accelerates.
CNBC and other outlets show how AI tools are enabling tiny firms to scale without large staffs, with Larridin’s measurement platform and solopreneur models illustrating rising AI-driven productivity and revenue channels.
Fal has secured a new funding round at a $16 billion valuation, doubling from April’s $8 billion mark. The Series E & F rounds total $1.65 billion, led by Sequoia previously and now backed by Nvidia, Sound Ventures and T. Rowe Price. Fal’s infrastructure supports AI inference, with revenue surpassing $800 million annually; margins remain thin due to compute costs, as peers in the space race to scale.
Google has set Gemini, Gemini Notebook, Gemini in Classroom and Google Search off by default for K–12 student accounts and said school administrators must explicitly enable them. Google has also said its Search features are "deeply integrated" with AI, so schools that block AI entirely will lose access to Google Search until they re-enable it.
WhiteMatter has built guardian agents to supervise AI agents within enterprises, a move amid rising concerns about agent security. The company says guardian agents monitor access and outcomes, scaling with new agents; competitors include Prompt Security, Lakera, and Protect AI. CrowdStrike is rolling out an AI kill switch to stop rogue agents, reflecting a broader push for internal AI safeguards.
Elon Musk has spent roughly $446m on Republican-aligned Super PACs in the 2026 US midterms, becoming the single largest donor, but his preferred candidates lost key races — including Texas — helping Democrats take control of Congress. Musk has remarked on X that his spending failed and joked about "space" data centres after results emerged.
Voters nationwide are rejecting most data-center measures in 2026, with Ohio emerging as a focal point in the data-center policy debate. Harwood, North Dakota, is the lone exception where a measure is defeated; energy costs and local infrastructure shape outcomes as campaigns frame data centers within broader economic concerns.
Google has rolled out controls that turn Gemini, Gemini Notebook and Search AI off by default for K–12 student accounts and let school administrators enable them. The policy has provoked backlash from parents, educators and politicians who say Google is pressuring schools and using its search dominance to force AI into classrooms.
Tencent Cloud is broadening its enterprise AI offerings with Hy4 and Hy4 Flash models while expanding WorkBuddy, its AI productivity agent. The company is targeting international markets, including the Middle East, and is building a large ecosystem of developers and partners as cloud regions grow to 25 with 70 availability zones.
A wave of AI agents is shifting shopping, work automation, and robotics. Retailers tailor product pages for bot discovery, firms test multi-agent collaboration, and startups push synthetic data for training physical AI models. The pace of change is accelerating as companies experiment with autonomous agents and human-in-the-loop safety measures.
Apple has removed the waitlist for Siri AI on iOS 27, enabling automatic downloads for eligible devices unless demand spikes. The feature, revealed at WWDC, now operates on a modern architecture and expands to more devices, with beta access and device compatibility required.
Anthropic has filed its S-1 ahead of a major IPO, presenting a life-sciences vision for AI. The company highlights Claude as a catalyst for rapid biological progress, backed by partnerships with pharma giants and a Bay Area biology lab. Industry skepticism remains about cures and clinical realities, even as AI accelerates drug discovery.
Anthropic has set a price range of $70–$78 for its shares in an IPO that could raise up to $64 billion and values the company at about $1.26–$1.41 trillion. The offering kicks off a roadshow amid volatile markets and concerns over AI spending.
Voters have backed tighter rules on AI datacenters in multiple states, forcing delays in planned capacity. Democrats have won key governor races, signaling increased governance over AI infrastructure and its economic impact. The results show broad political pressure to curb unchecked tech expansion ahead of a pivotal 2028 cycle.