What happened: TechCrunch reports a wave of users trying Claude after a series of OpenAI-related controversies, framing the shift as both a values-driven reaction and a practical search for a “safer” default assistant.
Why it matters: Switching assistants isn’t just about installing a new app — it’s about portability of context. The piece treats “memory” and chat-history exports as the real lock-in layer, and shows how people can move that personal context (carefully) to Claude.
Wider context: The story sits inside a broader trust and governance fight: model vendors are being judged not only on capability, but on who they’ll sell to and what guardrails they’ll accept — and those decisions can show up immediately as user growth, rankings, and churn.
Background: The guide walks through exporting ChatGPT data via Settings → Data Controls, enabling Claude’s Memory feature, and then (if desired) deleting ChatGPT account data rather than merely canceling a subscription.
Users are ditching ChatGPT for Claude. Here’s how to make the switch — TechCrunch
Singularity Soup Take: “Switching” is the easy part; the hard bit is deciding which parts of your life you actually want an AI to remember — portability should come with a little less hype and a lot more discipline about what gets saved.
Key Takeaways:
- Portability is the real feature: The article argues you can move the useful parts of your ChatGPT history — preferences, recurring topics, and instruction-style summaries — to Claude so it can respond more consistently without starting from zero.
- Export before you quit: TechCrunch points to ChatGPT’s built-in export flow (Data Controls → Export Data) that can produce text/JSON archives delivered by email, which may take time if you have a large history.
- Don’t paste raw logs: For importing into Claude, the recommended approach is to provide curated context or ask Claude to summarize exported files, then verify what it stored — reducing the risk of dumping noisy or sensitive material straight into memory.