Snap Cuts Jobs, China Blocks AI Chips, and Gemini Deep Research Hits Gmail

Week of May 2, 2026

Snap's decision to lay off 1,000 employees—16% of its workforce—marks a stark moment where AI transitions from augmenting work to replacing it. The company says AI now writes more than two-thirds of its new code, promising half-a-billion dollars in annual savings. That shift reverberates across tech as firms weigh AI-driven productivity against headcount.

Snap's AI-Powered Layoffs

Evan Spiegel announced the cut of roughly 1,000 jobs and the closure of over 300 open roles, citing "rapid advancements in artificial intelligence." Internal data shows AI-generated code now accounts for more than 65% of Snap's new software output, a jump that lets smaller teams maintain the same velocity. The restructuring is projected to deliver over $500 million in annualized cost savings by the second half of 2026, helping Snap push toward net-income profitability. Investors reacted positively, with the stock climbing 11% in pre-market trading on the news.

The layoffs underscore a broader trend: AI is no longer just a tool for augmentation; it is becoming a lever for operational redesign. Companies that once relied on large engineering squads are now rethinking headcount as models take over routine coding, testing, and even documentation. For Snap, the move also frees up capital to invest in new AI features—such as augmented-reality filters powered by generative models—potentially widening its competitive edge against rivals still scaling human teams.

China's Ban on Foreign AI Chips

In a sweeping policy move, China prohibited the use of foreign-made AI processors in state-owned data centers, a directive that appeared in the weekly summary and was echoed in multiple outlets. The ban targets high-performance GPUs and accelerators from companies like Nvidia and AMD, pushing domestic alternatives such as Huawei's Ascend series and indigenous designs from Biren Technology. Officials framed the rule as a safeguard for technological sovereignty, citing concerns over supply-chain security and the desire to accelerate home-grown chip development.

The restriction immediately sent ripples through the global semiconductor market. Nvidia's China-focused B300 server prices have already nearly doubled to about $1 million per unit as smuggling channels dry up, and analysts warn that the ban could compress foreign vendors' revenue streams in the world's largest AI-infrastructure market. At the same time, Chinese data-center operators are scrambling to qualify local chips, a process that may temporarily slow AI deployment but could ultimately bolster domestic semiconductor capacity.

For multinational tech companies operating in China, the ban presents both compliance challenges and strategic opportunities. While they must navigate the new restrictions on hardware procurement, some may find openings to partner with Chinese chip manufacturers or develop hybrid solutions that blend foreign and domestic technologies.