AI Threat Assessment · 26 May 2026

GitHub

Developer Platform
STILL BREATHING
2.8/ 10

“GitHub built the social network for code and somehow convinced every developer on Earth that git repositories needed infinite scroll and a follow button. They turned version control into a lifestyle brand, complete with green contribution squares that make grown engineers check their commit streaks like fitness influencers counting steps. The cruel irony is that they've become so essential to modern development that AI companies need GitHub to train the models that will automate away half of what gets committed to GitHub.”

Business Model
2.5
Automation Risk
4.0
Moat Strength
2.0
Adaptability
2.5
Need Survival
3.0
AI Threat Level
BUSINESS MODEL REPLACEABILITY

GitHub charges rent on the infrastructure every AI model depends on — the training data, the deployment targets, the collaboration workflows that turn prompts into production code. Copilot writes the code, but it still needs somewhere to live, and that somewhere writes Microsoft a very large check every month.

2.5
WORKFORCE AUTOMATION RISK

The platform that hosts Copilot is watching AI eliminate the need for junior developers, code reviewers, and DevOps specialists — which is to say, 60% of their user base. The remaining seniors will still need GitHub, but they'll need a lot fewer seats.

4.0
MOAT STRENGTH

Network effects are real when 100 million developers have their professional identities, contribution histories, and open source reputations locked in your platform. Switching costs aren't about the git repositories — those are portable — but about the career capital that lives in your GitHub profile.

2.0
AI ADAPTABILITY SIGNALS

They're shipping Spark, Models, and MCP Registry while doubling down on Copilot integration at every layer — the rare platform that's not just hosting AI development but becoming the infrastructure that AI development happens through. They're building the roads, not just watching the traffic.

2.5
WILL THE NEED SURVIVE AI?

Code collaboration survives even when AI writes most of the code — someone still needs to review, deploy, and maintain the AI's output. The question isn't whether developers will disappear, but whether they'll need as many seats and as much storage for what becomes a more supervisory workflow.

3.0
Verdict

“GitHub is the landlord in a neighbourhood where AI is simultaneously increasing property values and reducing the tenant population — more valuable per developer, fewer developers overall. The platform that democratised coding might end up being the platform that makes most coders redundant, which is probably the most GitHub thing that could happen.”

Scores are based on public information and AI analysis. This is an affectionate roast, not a financial assessment. The best companies use this as a mirror, not a verdict.

Roast another →