Why ‘boring’ VS Code keeps winning

Every few months, the developer tool hype machine finds a new hero. In 2023, it was GitHub Copilot, the AI pair programmer that made autocomplete feel like magic. In 2024, the vibe shifted to Cursor and the new class of AI-first editors. And now, at least on X, Google’s “agent-first” Antigravity is being pitched as the next inevitable thing.

Meanwhile, the model layer keeps whiplashing. First, everyone used ChatGPT. Then Gemini was catching up. Now, it seems Claude is the default brain for developers who want reasoning over speed.

So much churn, but churn doesn’t necessarily translate into sustained adoption. It turns out distribution beats novelty, especially once the enterprise shows up.

That is why, even in 2026, the gravitational center of day-to-day development still looks a lot like Microsoft, with Visual Studio Code as the workbench, GitHub as the workflow hub, and GitHub Copilot as the default assistant bolted onto both. Hence, the real question isn’t “Will AI replace the IDE?” but rather “Who owns the control plane when AI becomes part of the IDE?”

The stubborn persistence of VS Code

In all the talk about Cursor, Antigravity, etc., it’s easy to forget that VS Code’s popularity keeps increasing. According to a 2025 JetBrains’ survey of 24,534 developers, 85% of them reported they use AI tools, and 62% rely on at least one AI coding assistant/agent/editor. The locus of all that AI use? According to the Stack Overflow 2025 Developer Survey, VS Code sits at 75.9% among all respondents and 76.2% among professional developers. Compare this to 2024, when it was at 73.6%.

This isn’t collapse. It’s entrenchment.

Yes, Cursor shows up strongly at 17.9%, a massive leap for a newcomer. But here is the part that matters strategically. A lot of new AI editors are not replacing the VS Code ecosystem; they are riding it. Cursor’s own documentation highlights its seamless migration from VS Code because it is a fork of the VS Code code base. Google’s Antigravity is also a fork of VS Code that integrates Gemini 3, aimed at developers and enterprises who want agentic workflows in the editor.

Even when developers move to the hot new thing, the gravitational pull points back to the same platform primitives: extensions, keybindings, and repo integrations forged inside the VS Code universe. This is the “Intel Inside” problem for developer tools. Ecosystem is a hard habit to break.

GitHub Copilot isn’t going away

If you measure “winning” by how many developers actually touch the tool, GitHub Copilot is not a sideshow. GitHub CEO Thomas Dohmke recently reported 20 million GitHub Copilot users, up from 15 million just a quarter prior to that, while also noting that 90% of the Fortune 100 use the tool. That’s incredible growth, especially in light of alternatives mostly dominating the hype. Of course, this doesn’t mean Copilot is universally beloved. It’s not. But it is everywhere.

In the enterprise, procurement, compliance, and “it is already in the tool chain” matter more than vibes.

This scale feeds the real advantage: distribution. GitHub is the workflow. VS Code is the workbench. GitHub Copilot is the default assistant bolted onto both. As I have argued previously regarding Oracle’s converged database strategy, enterprises prefer integrated suites over best-of-breed fragmentation because integration reduces the complexity tax. Microsoft is applying that same converged strategy to the developer experience.

Keeping developer trust

If there is a threat to Microsoft’s dominance, it isn’t features. It’s trust. 2025 was a bad year for GitHub’s reputation, not just because of outages, but because of a growing perception that Microsoft is prioritizing AI adoption over developer agency.

We see this most clearly in the friction around opting out. In 2025, Microsoft and GitHub challenged developer trust by pushing GitHub Copilot deeper into core workflows without giving maintainers clean, reliable control over it. For example, two of the most upvoted GitHub Community threads in the prior 12 months were requests to block Copilot-generated issues and pull requests, and to fix the inability to disable automatic Copilot code reviews.

Beyond this friction, GitHub has made ecosystem-level shifts that feel like rug pulls to integrators. In a move that shocked many, they announced a hard sunset for GitHub Copilot Extensions built as GitHub Apps, blocking new creation after September 24, 2025, and enforcing full disablement by November 10, 2025. By explicitly telling developers this was a replacement rather than a migration as they pivoted to Model Context Protocol servers, GitHub violated the cardinal rule of “boring” infrastructure. Stability is supposed to be the feature, not API churn.

And just to round it out, GitHub Copilot’s security posture took a very public hit when researchers disclosed “CamoLeak,” a critical Copilot Chat vulnerability that could exfiltrate secrets and private code from private repos via prompt injection and a Content Security Policy bypass, which GitHub mitigated in part by disabling image rendering in Copilot Chat. Put those together and the trust problem is not that AI exists, it’s the perception that GitHub Copilot is becoming unavoidable infrastructure, while simultaneously being subject to churn and occasional sharp edges that are hard to justify when the product is supposed to be the boring, dependable layer.

Which maybe, just maybe, opens the door for Google.

Can Google sustain the hype?

Antigravity is a legitimate technical marvel. It represents a shift to “agent-first” development, where you delegate high-level tasks to Gemini 3 agents that run across the editor, terminal, and browser. It does this while borrowing VS Code’s familiarity.

But familiarity can be good and bad. Google’s historical weakness is not innovation; it is commitment. Developers (and the CTOs who approve their tools) have learned to fear the “killed by Google” roulette wheel. For an enterprise to rip out VS Code for Antigravity, they need to believe Antigravity will exist in 2030. Google’s track record makes that a hard bet to place. That said, lately Google has become much better at being boring and has become essential enterprise infrastructure with Google Cloud. If Google is able to manage AI security concerns better than Microsoft, it could become the new “boring.”

After all, the winners in this market won’t be the companies with the most hype; they’ll be the ones that can take a chaotic model market and turn it into a calm, governed, low-friction developer experience.

Microsoft has done that for decades, which puts it in a strong, but not unassailable, position. For developers, Microsoft owns the default workbench (VS Code), the default workflow hub (GitHub), and the enterprise rails that turn experimentation into standardization. Even when developers “switch” to Antigravity, they are often just moving to a different room in Microsoft’s house, as it were.

This doesn’t mean Microsoft will keep winning, but it does mean they’ve set the “boring” bar high.