The AI Productivity Lie(parand.com)
parand.com
The AI Productivity Lie
https://www.parand.com/the-ai-productivity-lie.html
3 comments
> If so, why is the pace of innovation and updates in their AI development products (Codex, Claude Code, Antigravity) so pedestrian?
Claude seems to be pumping out features fairly rapidly?
Claude seems to be pumping out features fairly rapidly?
This paragraph is very confusing to me because it’s so easy to… check and falsify?
Anthropic has released ~90 minor versions over the course of just shy of two months (2.1.9 from February 19; 2.1.100 from today). Over the course of these versions, the following features were added, among others:
- Task system
- in-chatbox bash autocomplete
- plugin overhaul
- custom keybindings
- PR review mode
- GH CLI auto-integration
- Agent Teams
- Auto-memory
- MCP overhaul
- Remote control
- /simplify
- Batch tasks
- text-to-speech
- hook overhaul
- looped/scheduled tasks
- Worktree management
- MCP elicitation
- MCP channels
- bare mode
- deferred hook actions
- Onboarding flow, /powerup command
- Monitor tool
And IDE extensions and the Agent SDK have been worked on during that time as well.
There’s an entire different argument here about whether that sort of release cadence paired with the growing list of Issues and the rather revealing leak from the other day is something that’s desirable. But to call the pace of releases standard or imply the changelogs are all mundane is just straight-up false.