Amazon Drives AI Coding Adoption with Mandatory In-House Tool Kiro
Amazon is aggressively promoting the use of AI-assisted coding as a standard practice across its entire organization. According to a report from the Financial Times, the company has set a clear target: 80 percent of its developers should utilize AI for coding tasks at least once every week. However, Amazon is not allowing its engineers to freely choose their tools. Instead, it is strongly encouraging them to adopt Kiro, its proprietary AI coding assistant launched in July 2025, while actively discouraging the use of alternative solutions.
Internal Memo Reveals Strict Policy Against Third-Party AI Tools
An internal memo reviewed by Reuters in November detailed Amazon's firm stance. Signed by two senior vice presidents—Peter DeSantis of AWS utility computing and Dave Treadwell of eCommerce Foundation—the document explicitly stated, "We do not plan to support additional third-party AI development tools." It designated Kiro as Amazon's "recommended AI-native development tool." Following a six-month review, OpenAI's Codex was labeled as "Do Not Use." Anthropic's Claude Code initially received the same designation, but this was later reversed. A spokesperson informed Business Insider that approximately 70 percent of Amazon's software engineers used Kiro at least once in January, with the company aiming to rapidly increase this percentage.
Employee Resistance and Internal Conflicts Emerge
Not all Amazon employees are supportive of this directive. Business Insider reported that around 1,500 Amazon employees endorsed the formal adoption of Claude Code in an internal forum discussion. Many argued that Claude Code is superior, with one employee commenting, "A tool that can't keep pace with rivals offers no real innovation," and adding that Kiro's "only survival mechanism becomes forced adoption rather than genuine value." The frustration is particularly acute among AWS sales engineers, who are responsible for selling Claude Code to customers via Amazon's Bedrock platform but are prohibited from using it for their own production work. One engineer expressed concern, writing, "Customers will ask why they should trust or use a tool that we did not approve for internal use." Amazon maintains that there is no outright ban on Claude Code but applies "stricter requirements" for production tools.
Complexities Amidst Major AI Investments and Workforce Changes
This situation is complicated by Amazon's significant financial stakes in rival AI companies. The company has invested $8 billion in Anthropic, the creator of Claude Code, and secured a $38 billion cloud computing deal with OpenAI, whose Codex tool is also restricted internally. The push for Kiro is part of a broader AI transformation under CEO Andy Jassy, who has indicated that AI-driven efficiency gains will lead to a reduction in Amazon's corporate workforce over time. The Financial Times reported that Amazon has cut 30,000 corporate roles since October 2025—its largest reduction ever—while committing a record $200 billion in capital expenditure this year, primarily directed toward AI infrastructure and data centers.
Engineers Express Preference for External AI Models
Despite Amazon's heavy investment in its own AI tools, some engineers remain unconvinced. Several told the Financial Times that they prefer Anthropic's Claude over Amazon's Nova models for coding tasks. One AWS engineer bluntly stated to FT, "I didn't even know we had a model," highlighting a potential disconnect between corporate strategy and on-the-ground developer sentiment.
