Amazon confirmed on July 22nd that it had cut some positions within its AGI organization, stating that the company is still working on building large AI models, which it considers one of the most critical tasks. According to Reuters' report, Amazon explained this adjustment as a refocusing of resources on the areas most important to customers' future.
The exact number of layoffs was not disclosed, and this move should not be directly interpreted as Amazon abandoning AGI. Rather, it seems to bring to the forefront the contradictions in Amazon's AI narrative: while big tech continues to invest in AI infrastructure at the multi-billion-dollar level, the team closest to the long-term AI ambitions is undergoing organizational shrinkage.
For investors, the question is not how many people Amazon laid off but rather a reevaluation of AWS's AI valuation anchor. In the past, the market was willing to pay a premium for big tech AI investments, assuming that the stronger the model, the greater the future revenue. The more pressing issue now is when these investments will translate into customer payments, cloud revenue, and improved profit margins.
AGI (Artificial General Intelligence) can be understood as a long-term goal that AI, like humans, can learn across domains and solve problems. It represents a vision for the future but may not immediately translate into revenue. What AWS needs is to package AI capabilities into services that enterprises can buy, use, and customize now.
As AI Investment Grows, Organizational Trade-offs Harden
The key to this round of layoffs is not whether Amazon will continue to work on models. The official statement has already set the boundary: large models remain a priority, but resources need to be allocated to projects that matter most to customers and have the highest priority.
The organizational moves reveal that AI investment has not stopped, but the tolerance for error in investments is decreasing. In December 2025, Amazon reshuffled its AI-related leadership, with Andy Jassy announcing Peter DeSantis's responsibility for a new organization focused on AI models, chips, and quantum computing. Rohit Prasad departed by the end of 2025, with Pieter Abbeel taking charge of cutting-edge model research within AGI. Reuters' report also mentioned that David Luan, head of the AGI Lab, left in February 2026.
Viewed together, these changes show that Amazon is not withdrawing from the AI arms race, but rather reordering its internal investment portfolio. Long-term research still holds narrative value, but projects closer to customers, revenue, and productization are receiving higher priority.
This is also the backdrop to a broader trend in big tech AI. Over the past two years, the market primarily traded on who dared to spend, who had the computing power, and who had the models. Now, capital expenditure itself is no longer scarce enough, and investors are starting to ask about returns: whether model teams, chips, data centers, and talent can ultimately translate into revenue.
Nova Forge Provides Commercialization Leverage
To understand this adjustment, one needs to look at AWS's Nova Forge, released during AWS re:Invent in December 2025. It is not an ordinary chatbot but a service that helps businesses train custom models.
In the traditional path, if a company wanted a cutting-edge model tailored to its industry, they either had to train from scratch, which was very costly, or fine-tune a pre-trained model with limited capabilities and controllability. The idea behind Nova Forge is to allow customers to start from checkpoints in the Amazon Nova model training process (mid-training archives), mix in their own data and Amazon-curated datasets at different training stages.
Amazon calls this open training. Simply put, companies don't have to build a large model from scratch but can start with a model base that Amazon has already trained to a certain stage and inject their industry knowledge early on. This way, they can inherit the foundational capabilities and more easily develop domain expertise.
This path is crucial for AWS as it attempts to turn model capabilities into cloud service products. Customers are not just calling an API for a model but are training, hosting, deploying, and optimizing their models on AWS. If the product succeeds, it could lead to compute consumption, platform stickiness, and ongoing operational revenue.
However, existing information does not prove that the dismantled AGI resources have already shifted to Nova Forge. A more cautious assessment is that the AGI organization adjustment and customer-oriented products like Nova Forge have appeared simultaneously, demonstrating Amazon's inclination to increase the emphasis on commercialization projects.
AWS Shifts Competitive Focus to Customer Customization
Amazon's position in the foundational model competition has always been somewhat unique. It both develops Nova and invests in Anthropic while also maintaining AWS's neutrality as a cloud platform and the model ecosystem.
This determines that AWS may not necessarily win solely by having the world's most powerful models. For enterprise customers, the model rankings matter, but they are not the only criteria. The more practical questions are whether they can access internal enterprise data, meet security and compliance requirements, reduce training costs, and seamlessly integrate with existing cloud services.
The very essence of Nova Forge is to challenge this competitive logic. It shifts the battlefield from general model capability ranking to whether enterprises can train their own models at a lower cost. If this approach proves successful, AWS can embed AI revenue into its core cloud computing business, rather than solely betting on a consumer-grade AI product.
This also explains why Amazon is simultaneously holding on to the AGI narrative while downsizing certain positions. The former maintains long-term technological ambition, while the latter forces teams to reallocate resources towards directions that are more easily validated by customer demand.
For AMZN, the market will ultimately not only focus on whether Amazon has an AGI team or not. More importantly, it will hinge on whether AWS can demonstrate that AI services increase customer spending, enhance stickiness, and do not significantly drag down profit margins.
Orders and Profit Margins Will Provide the Answer
This round of layoffs could easily be misinterpreted in two extremes: either as Amazon's AI failure or as an insignificant routine optimization. Existing information does not support such conclusions.
A more reasonable assessment is that Amazon is still engaged in the AI arms race, but internal budgets and talent allocations are shifting towards directions that can be sold to customers. This shift is significant for investors because AMZN's AI premium will increasingly rely on the commercialization outcomes of AWS, rather than solely on model narratives.
The validation points will revolve around specific criteria. Whether Nova Forge can onboard real enterprise customers, whether customers are willing to continue paying, and whether the trained models are more cost-effective than regular fine-tuning will all determine its effectiveness as a product.
Another variable is talent attrition. If the AGI organization adjustment is merely about optimizing non-critical positions, the impact will be limited. However, if core research and engineering talents depart, Amazon's long-term competitiveness in proprietary modeling will be compromised. The tension between official statements and organizational realities will ultimately be absorbed by product adoption rates, AWS AI revenue, and capital expenditure returns in subsequent financial reports.
Welcome to join the official BlockBeats community:
Telegram Subscription Group: https://t.me/theblockbeats
Telegram Discussion Group: https://t.me/BlockBeats_App
Official Twitter Account: https://twitter.com/BlockBeatsAsia
