AI coding models
Topic
What experts have said about AI coding models
9 statements · 5 positive · 3 negative · 1 neutral
Perfectly coding AI models would find every exploitable software vulnerability.
“if these models can effectively code perfectly, then there is no exploit that it's not going to find”
Open the episode · GPT-6 Hits AGI? Tech Euphoria 2.0, SF Mansion Shortage, NYC Bans AI in Schools & Venezuela Oil DealListen at 21:10
AI models can build essentially any software feature from scratch.
“I generally believe now that starting from scratch, if you talk to these models, ours, anybody else's, really, you can stand up whatever feature you want”
Open the episode · OpenAI Codex lead on the new shape of product work | Andrew AmbrosinoListen at 3:09
Correct code generation accelerates AI research more directly than design capability.
“the model being able to write correct code would accelerate research in a way that you can't really make the same case for design.”
Open the episode · OpenAI Codex lead on the new shape of product work | Andrew AmbrosinoListen at 13:09
Current AI models usually increase software complexity.
“One thing that I think all models suffer with right now is just they, they usually increase complexity.”
Open the episode · OpenAI Codex lead on the new shape of product work | Andrew AmbrosinoListen at 40:07
New AI models outperform elite programmers at software coding.
“these new models are now better at coding than I am”
Open the episode · #2501 - Marc AndreessenListen at 2:02:22
Current AI coding models perform hyperparameter optimization effectively.
“the models can do a very good job of doing hyperparameter optimization”
Open the episode · Eric Jang – Building AlphaGo from scratchListen at 2:22:54
David Friedberg predicts developers will retain preferences over which AI models write code.
“the developers are going to want to continue to have at least some choice on what's actually writing the code for them.”
Open the episode · SpaceX-Cursor Deal, SaaS Debt Bomb, New Apple CEO, SPLC Indictment, Colon Cancer SpikeListen at 11:50
Chamath says current AI models cannot reliably build enterprise-grade software.
“the long horizon ability for any of these models to actually build enterprise grade software is still S H I T”
Open the episode · Anthropic's $30B Ramp, Mythos Doomsday, OpenClaw Ankled, Iran War Ceasefire, Israel's InfluenceListen at 33:37
Coding models currently provide roughly 15–20% total-factor productivity improvement.
“the coding models give maybe, I don't know, a, like 15, maybe 20% total factor speed up”
Open the episode · Dario Amodei — The highest-stakes financial model in historyListen at 37:36
Statements are attributed to the speaker as said on the episode and reflect their view at the time, not PodLume's. They are not advice.