Link to detailed paper

Leave a comment

Link to document version with article summaries

Here are the weekly AI news:

Anthropic:

• Anthropic discloses Claude Opus 4.7, Mythos 5, internal model hacked three organizations in April during cyber testing; 141,000 runs reviewed theverge

OpenAI:

• July 30: OpenAI cuts GPT-5.6 Luna 80% to $0.20/$1.20, Terra 20% to $2/$12, adds Sol fast mode 2.5x speed at 2x price. startuptalky

NVidia:

• Nvidia and SK Group unveil $500B AI initiative with 2GW data centre using Vera Rubin chips and high bandwidth memory HBM4 by 2027 livemint

Investment:

• Nscale acquires Anyscale for $1.65B to vertically integrate AI workload management with its neocloud infrastructure. techcrunch

• CXMT’s $8.6B semiconductor IPO on July 27, 2026, surges 531%, becomes China’s top A-share firm, driven by AI memory demand. technode

• Nvidia’s reported $250B financing for OpenAI’s 10-gigawatt data center project shifts AI leadership to infrastructure, capital, and energy. devdiscourse

Automated Driving:

• FREENOW partners with Baidu Apollo Go to test Apollo RT6 robotaxis in London’s Brent borough, public rides begin 2027. technode

China:

• Pboc launches national tech finance data catalogue with 26 indicators across 8 categories on July 29, 2026 technologynewschina

• China’s AI models train at one-tenth cost, API at 10-20% of rivals, with 20-40% gross margins. technode

• Open-weight Moonshot Kimi K3 nearly matches GPT-5.6; Microsoft coalition backs open models over proprietary. zdnet

• Model distillation becomes latest US-China AI battleground as Anthropic accuses DeepSeek, Moonshot, MiniMax of extracting Claude capabilities; OpenAI also detects. livemint

Europe:

• EU opens €30B tender for up to seven AI gigafactories, bids due November 12, 2026, awards early 2027. tagesschau

• Google signs EU AI Act code of practice with SynthID partnerships, warns regulatory complexity contradicts EU competitiveness goals. blog

Germany:

• Munich court rules Suno infringed GEMA copyrights, rejecting text-and-data-mining defense; injunction immediately enforceable, damages later. techtimes

Hardware:

• 1950 billion in forward HBM (high bandwidth memory) contracts lock AI accelerator supply through 2030, leaving unsecured firms facing years-long shortages. techtimes

Agents:

• In 2026, AI automation market at $169.46B dwarfs RPA’s $35.27B, but enterprises need both in intelligent automation layers. tekrevol

Large Industry-Specific AI Models (LIMs):
• SAP’s July 27, 2026 tabular AI foundation model with table-native transformer and in-context learning reduces retraining for autonomous enterprise. towardsai

Healthcare:
• AI-driven drug discovery hits data wall, needs negative data and autonomous labs to close loop by 2026. technologyreview

Adoption & Transformation:
• AI governance platforms market emerges with dedicated tools, cloud integration, and growing agentic AI controls in 2026. techtarget

Regulation & Government:
• Trump administration bans new foreign-made humanoid robots, robot dogs, and power inverters citing national security; 15,000 humanoids shipped in 2025 techcrunch

• June 2026 federal restrictions on Anthropic and OpenAI mandate six minimum conditions for frontier AI governance. ari

• Over 1,100 AI workers urge U.S. to develop governance tools for slowing automated recursive self-improvement. businessinsider

Curated with AI from 1109 news reports. Article summaries are in the attachment for viewing in a browser (not email reader).
I, AI and underlying news can make mistakes. Check important info.

Yours,
Robert

Subscribe now

Abstract

“Practical AGI” (artificial general intelligence) has arrived and deployment is moving from pilot to production among the leaders. Now AI turbo-charges competition: every rival can cut cost and build new products fast. The advantage goes to whoever moves first. The lead is won inside the 2026–2028 window.

For businesses, this is broken down in 6 chapters:

• Rebuild business functions AI-first. This resets the cost base every rival competes on: operating cost reductions of ~21% are possible now, rising to ~48% as robotics and agents mature. Start with quick-win copilots that pay back in under a year.

• Lead your sector by reinventing the product. The deepest AI gains create products that reshape whole industries: Robotaxi rides triple a year and the first AI-designed medicines reach the clinic. First movers out-innovate faster than rivals can copy. Build deep, narrow AI on your own data, not a generic copilot.

• Close the AI implementation gap. The limit is execution, not the AI offering. 88% of firms run AI, yet ~95% of pilots show no profit in six months. Readiness decides the outcome, and mid-market firms can build it. One company AI platform, wired to your data and grading its own output, scales proven wins to return fast.

• Compete in the agent-mediated economy: the platform that owns the agent owns the customer, because it holds the customer’s data and context. AI is expected to mediate $20.9B of US retail in 2026. Ensure your best presence in the new agent platforms. Build a direct channel you own, so customer memory stays yours.

• Quantify and drive the bottom-line impact. The prize is huge but mostly not the firm’s to keep long term. Competition eventually converts most of the cost reduction into lower prices, and by ~20 years the average firm earns near-zero additional economic profit. Speed wins the transition and justifies early investments. The moat wins the end: brand, proprietary data, customer memory, or patents.

• Turn an understanding of the other players into business advantage: the AI firms you buy from, the people you employ and sell to, the governments that set your rules, and the world system. Competition within each player drives AI progress. This compresses a decade of innovation into the next 10 quarters, the remaining 2026–2028 window, where the lead is won.

Get the full paper here as HTML or PDF.

Feedback is highly welcome — leave a comment below.

Leave a comment