When you purchase through links on our site, we may earn an affiliate commission. This doesn’t affect our editorial independence.
Nowadays, it’s easier for a camel to enter through the eyes of a needle than to find programmers who aren’t using AI coding assistants in some professional capacity, especially for repetitive tasks. However, those who refused to follow the Coinbase AI coding mandate now face the music after Coinbase bought enterprise licenses for GitHub Copilot and Cursor.
According to CEO Brian Armstrong, who revealed this week on John Collison’s “Cheeky Pint” podcast, that engineers were dismissed under the Coinbase AI coding mandate. After securing licenses for every engineer, some notified Armstrong that adoption would be slow, suggesting months before widespread AI assistant adoption.
Contrarily, Armstrong disagreed with that view. “I went rogue,” he said while describing his bold decision to push Brian Armstrong’s AI coding policy. Additionally, he posted a direct mandate in the company’s leading engineering Slack channel, ordering engineers to onboard their AI assistant accounts quickly.
In Case You Missed It:
OpenAI Unveils GPT-5 to Shape the Future of ChatGPT
Google Rolls Out ‘AI in Search Mode’ for Nigeria, Ghana and South Africa
Armstrong noted, “I said, ‘AI is important. We need you to learn it and be on onboard immediately.” The company gave engineers until the week’s end to onboard. If not, Armstrong threatened a Saturday meeting to understand their resistance.
Interestingly, at the meeting, some engineers had legitimate excuses. Some claimed to have just returned from vacation, while others resisted AI coding assistants.
Coinbase AI Coding Mandate: Engineers Fired Over AI Tools Despite Objections
Moving forward, Armstrong said a few employees lacked valid reasons for ignoring the Coinbase AI coding mandate. Bluntly, he said those engineers without excuses were fired. “I jumped on this call on Saturday, and there were people who had not done it,” Armstrong explained.
Some engineers were just back from travel. However, others had refused AI assistant adoption. He dismissed the latter group. Armstrong admitted that it was a heavy-handed approach, and many within the company disliked Brian Armstrong’s AI coding policy.
It is noteworthy that while many engineers weren’t fired over AI tools, Armstrong emphasized that it sent a clear message regarding the Coinbase AI coding mandate. Additionally, the CEO pointed out that engineers cannot ignore AI’s influence. He made it clear that AI coding assistants are no longer optional.
Since the mass disengagement of engineers, Armstrong has leaned further into training. He has influenced the company to hold monthly sessions where teams demonstrate advanced AI assistant adoption strategies.
Broader Debate on AI-Coded Codebases Emerges
Interestingly, podcast host Collison, president of Stripe, questioned how much companies should ultimately rely on AI-generated code. Nevertheless, Armstrong’s response was, “It’s clear that it is helpful to have AI coding assistants.” “It’s unclear how to manage AI-coded codebases.”
Interestingly, Armstrong agreed by acknowledging the challenge of balancing productivity gains with long-term maintainability under the Coinbase AI coding mandate for all engineers.
It is, however, notable that even OpenAI struggled earlier. A former engineer described its central repository as “a dumping ground.” Management at OpenAI has since dedicated engineering resources to cleaning up AI-generated code and improving internal systems for long-term sustainability.
As this report comes to a close here, the context contributes to a larger debate as to how Coinbase fired engineers over AI tools. Do companies now weigh innovation against cultural resistance and productivity?
Nevertheless, as rhetoric as this question appears, the Coinbase AI coding mandate might inspire other firms to force AI assistant adoption. However, balancing enforcement and creativity remains an unresolved issue.
