Published on LinkedIn and amitabhapte.com  on 5th April 2026

Three forces are running simultaneously through the AI landscape right now: a scramble for physical control of the underlying infrastructure, a repricing of how AI-driven value gets negotiated, and a fracture in how the world governs it. This week’s signals belong to all three.

1. The Race to Own the Substrate

New IDC data reviewed by Reuters shows Chinese GPU and AI chip makers captured 41% of China’s AI accelerator market in 2025. Huawei alone shipped around 812,000 chips, with Alibaba’s T-Head and Baidu’s Kunlunxin growing behind it. Nvidia still leads at 55%, but the retreat is real. 

The same instinct is driving infrastructure investment across Asia. Airtel raised $1 billion for its data centre arm Nxtra from Carlyle, Alpha Wave, and Anchorage Capital, targeting a scale-up to 1GW of capacity in India, a country that already has Google’s $15 billion data centre commitment and a 20-year tax holiday for hyperscalers. Microsoft’s $10 billion commitment to Japan through 2029 is structured similarly: AI infrastructure, national cybersecurity cooperation, and data processed inside Japan’s borders. These are sovereignty arrangements as much as commercial deals. 

2. AI Is Changing the Terms of Access

In enterprise software, ServiceNow’s CEO Bill McDermott has repositioned the company around a single argument: AI models identify problems but struggle to execute reliably across governed, auditable enterprise workflows. ServiceNow owns the last-mile execution layer, and its Now Assist product is tracking toward a $1 billion annual run rate. The model shift underneath it is structural: from per-seat licensing to outcome-based pricing, what McDermott calls digital labour. 

In capital markets, the dynamic is rawer. SpaceX is targeting a valuation above $2 trillion in what could be the largest IPO in history. According to the New York Times, Musk has required the lead banks, Morgan Stanley, Goldman Sachs, JPMorgan, Bank of America, and Citigroup, to purchase Grok subscriptions as a condition of the mandate. Some have agreed to spend tens of millions per year. AI adoption bundled into the price of deal access is a new distribution model. 

3. AI Governance Is Becoming a Procurement Reality

Several jurisdictions are now introducing procurement standards that require AI vendors to demonstrate safety and privacy safeguards before accessing public sector contracts. California’s Executive Order N-5-26 is a recent example: companies seeking state contracts must disclose safeguards against harmful content, bias, and civil rights violations, with agencies conducting their own independent assessments. The practical result is a growing patchwork of standards that enterprise technology teams will have to navigate across geographies. California’s procurement rules have a history of becoming de facto global benchmarks, much as GDPR did from a single jurisdiction. Any organisation selling AI products into regulated markets or the public sector should treat this as an active compliance question, not a future one.

4. Japan: Automation as Survival, Not Disruption

TechCrunch reported this week on Japan’s push into physical AI, driven by demographic emergency rather than efficiency ambition. With over 28% of the population above 65 and a working-age population contracting annually, Japan is deploying robots to fill positions that cannot be staffed. The government has committed $6.3 billion and is targeting 30% of the global physical AI market by 2040. The industry signal is unambiguous: customer-paid deployments, full-shift operation, measurable productivity. Japan is the clearest preview of what automation looks like when labour scarcity, not cost reduction, is the driver. 

My Takeaway This Weekend

The infrastructure layer of AI is being claimed physically and politically. The business models above it are repricing around outcomes. The governance frameworks meant to contain it are fracturing. And the labour markets it will reshape are already showing, in Japan, what comes next. AI strategy, infrastructure strategy, talent strategy, and governance strategy are now the same conversation. Running them separately is how organisations fall behind without noticing.