Amazon cuts jobs in its own general AI division: not even those building it escape the adjustment

By TradingView (GuruFocus) · July 23, 2026.

Amazon has cut jobs within its artificial general intelligence group (AGI: a hypothetical system that would match or surpass human intelligence and operate autonomously, a goal pursued by several of the big tech companies). The company has not disclosed how many people have been affected, according to TradingView, citing a statement from an Amazon spokesperson. It is the latest in a series of smaller cuts since the company eliminated 16,000 positions across the company in January.

The spokesperson explained that Amazon has been building large AI models for years and that this work remains "one of the most important" for the company, but that it is "sharpening its focus on the initiatives that matter most to customers." In its own wording, that reprioritization involves "difficult decisions," including eliminating some positions within parts of the AGI organization. The stock (AMZN) fell 1.35% intraday after the news broke. According to crn.com, the confirmation comes just before Amazon reports its second-quarter results, adding pressure to the market's reading of the real cost of its AI bets.

The figure itself is modest—there is no headcount for the layoffs nor a breakdown by team—but the symbolism is powerful: the very lab Amazon devotes to pursuing artificial general intelligence is not immune to the cuts that this same technology, in theory, promises to prevent across the rest of the company. Amazon is simultaneously running several AI fronts (Alexa+, AWS's Nova models, its investment in Anthropic, plus this AGI group), and a selective cut within AGI points more to a reallocation of resources among those bets than to an abandonment of the underlying ambition. The company is not saying that general AI is a dead end; it is saying that not everything labeled "AGI" two years ago survives today's review of priorities.

This episode fits with something we have been documenting at Zendoric for months: the impact of AI on employment does not respect clean corporate boundaries. Not even the teams building the models are safe from the efficiency logic that those same models make it possible to apply to the rest of the organization. Andy Jassy, Amazon's chief executive, warned back in 2025 that he expected the company's corporate workforce to shrink over time due to generative AI productivity gains; these AGI cuts are, in a way, that logic applied recursively to itself.

Our reading: this is not a sign that the race toward general AI is cooling—the big tech companies continue to invest record sums in compute and talent for that goal—but rather that the building phase is maturing into one of consolidation, in which projects with real product traction are funded and those without it are pruned. It is exactly the kind of short-term friction we anticipated as inevitable: restructuring is not reaching only call centers or administrative support, it is also reaching the AI labs themselves. The difference from other cuts is that here the underlying message—continuing to invest in large models because "the work remains among the most important"—supports the thesis that ambition is not fading, only being reorganized. The transition toward the abundance we defend as a long-term horizon does not exclude these shakeups; it includes them, and it is worth not confusing a headcount adjustment with a signal about the technological ceiling of general AI.