Bengaluru-based C2i Semiconductors, building a “grid-to-GPU” power system for datacenters, raised a $15M Series A led by Peak XV, bringing total funding to $19M
https://techcrunch.com/2026/02/15/as-ai-data-centers-hit-power-limits-peak-xv-backs-indian-startup-c2i-to-fix-the-bottleneck/
The AI boom’s power problem is no longer hypothetical; C2i’s “grid-to-GPU” approach exposes a fundamental flaw in the current datacenter model—wasteful energy conversion and transmission inefficiencies. Yet, pouring $15M into an Indian startup signals the West’s desperation to outsource not just manufacturing but also critical infrastructure innovation. This raises questions about technology sovereignty and whether dependency on emerging markets for core datacenter power tech might backfire amid geopolitical tensions.
Trump administration reaches a trade deal to lower Taiwan’s tariff barriers
https://apnews.com/article/trump-taiwan-china-trade-deal-2b1743397ba33010463d41132b75ce53
A trade deal lowering Taiwan’s tariffs sounds like free market progress but masks a strategic pivot: the U.S. under Trump is quietly tightening economic and tech alliances with Taiwan to counter China’s semiconductor ambitions. While framed as tariff relief, this deal effectively deepens supply chain entanglement with Taipei, risking escalation in cross-strait conflicts. Investors betting on de-risking China exposure should beware that Taiwan’s semiconductor sector remains a geopolitical powder keg.
OpenAI sidesteps Nvidia with unusually fast coding model on plate-sized chips
https://arstechnica.com/ai/2026/02/openai-sidesteps-nvidia-with-unusually-fast-coding-model-on-plate-sized-chips/
OpenAI’s move to bypass Nvidia’s GPU hegemony by deploying “plate-sized” custom silicon signals a tectonic shift in AI hardware dependency. But the mainstream misses the catch: this chip may solve speed but not the broader supply chain fragility tied to rare earths and fabrication capacity. The race to sidestep Nvidia’s GPUs is not just about performance—it’s an existential gamble on diversifying critical supply amid rising Sino-American tech decoupling.
Keir Starmer says no platform “should get a free pass”, as the UK moves to tighten online safety laws covering AI chatbots and social media, to protect children
https://www.ft.com/content/15917aa4-2d40-49be-85c3-da395b16e7f1
The UK’s tightening grip on AI and social media platforms under the guise of protecting children is less about safety and more about control. The “no free pass” mantra risks stifling innovation and driving AI development underground or offshore. By focusing heavily on regulation, the UK might inadvertently cede AI leadership to less constrained jurisdictions, thus undermining its own tech sovereignty ambitions.
Anthropic opens a Bengaluru office, its second in Asia, says its India run rate revenue doubled since October, and curates training data for 10 Indic languages
https://www.anthropic.com/news/bengaluru-office-partnerships-across-india
Anthropic’s expansion in India and focus on Indic languages highlights a critical but overlooked battleground in AI: linguistic and cultural diversity. However, rapid scaling in a market with rudimentary AI ethics frameworks and regulatory flux risks data privacy breaches and potential misuse. Global AI firms rushing into India must navigate not just opportunity but a minefield of governance gaps that could boomerang politically and reputationally.
India-based AI cloud startup Neysa plans to raise up to $600M in equity led by Blackstone, plus $600M in debt, to deploy 20K+ GPUs in India for AI training
https://www.bloomberg.com/news/articles/2026-02-16/blackstone-to-lead-600-million-investment-in-ai-firm-neysa
Blackstone’s massive financial backing of Neysa underscores the speculative frenzy around AI infrastructure in India, but stacking $1.2 billion in equity and debt into GPU deployment signals a dangerously leveraged bet on unproven market maturity. The focus on GPU scale alone ignores persistent challenges: power grid reliability, cooling infrastructure, and a talent crunch that could limit returns. This extravagance might inflate an unsustainable bubble in India’s AI cloud sector.
Sources: Hacker News, Techmeme, AP News, Ars Technica | Compiled 2026-02-16