In a recently released annual letter to shareholders, Amazon CEO Andy Jassy highlighted the company's significant investment in artificial intelligence (AI). He noted that while the development and deployment of AI currently remain expensive, the cost of using AI is expected to decrease significantly as technology advances.

Chips Technology (3)

Image Source Note: Image generated by AI, image licensing provider Midjourney

Jassy revealed that Amazon plans to invest up to $100 billion in capital expenditures by 2025, the majority of which will be allocated to AI-related projects. These investments will include the construction of data centers, upgrades to network equipment, the development of AI hardware, and improvements to generative AI service capabilities. Currently, Amazon has launched several AI products, including its self-developed large language model Nova series, its self-developed AI chip Trainium, a shopping chatbot for consumers, and the Bedrock platform serving third-party models. Additionally, Amazon has comprehensively upgraded its smart assistant Alexa with generative AI capabilities.

Since taking over as CEO in 2021, Jassy has been committed to streamlining Amazon's operations and controlling costs. According to reports, Amazon has laid off over 27,000 employees between 2022 and 2023. Looking ahead, Amazon plans to continue implementing smaller-scale layoffs and has already shut down several less profitable projects, such as the "try before you buy" clothing service and the offline express delivery store pilot program.

In addition to layoffs and project adjustments, Jassy has established an internal channel called the "anti-bureaucracy mailbox" to gather employee feedback. To date, this mailbox has received nearly 1,000 emails, resulting in over 375 optimizations to organizational processes and workflows. This approach reflects Amazon's efforts to improve internal efficiency and employee satisfaction.

Key Highlights:

🌟 Amazon plans to invest $100 billion by 2025, primarily in AI projects, including self-developed hardware and services.

💻 The launch of the self-developed Trainium chip and Nova series large language model aims to reduce AI usage costs.

✉️ The "anti-bureaucracy mailbox" collects employee feedback to help optimize internal company processes.