The Good and Bad of AI Coding: Amazon Shuts Down, Autoresearch, Claude Code Review, Lovable
March 12, 2026
AI Summary
5 min readAmazon’s entire storefront crashed for six hours last week after a junior developer submitted AI-generated code that brought the platform down. The outage cost the company millions, and it was the latest in a series of failures tied to Amazon’s aggressive push to have AI write 80% of its code by the end of 2026. Meanwhile, Anthropic—the creator of Claude Code, which Amazon uses—has been suffering its own outages, even as it releases a new paid product to review the code its own model generates. OpenAI is countering with a cheaper alternative. And a new open-source experiment from AI pioneer Andrej Karpathy hints at a future where models improve themselves autonomously. This episode of the Limitless Podcast walks through the tension between AI coding’s explosive demand and its growing pains.
The Amazon Outage: A Single Line of AI Code, Billions in Damage
Continue reading the full summary in the app — free to try.
Read Full Summary →Free • No credit card required
Never miss an episode of Limitless Podcast
Get every new episode summarized in your inbox — free, ~5 minutes to read.
No spam. Unsubscribe anytime.
What you'll learn
- 1 (00:00) **Amazon Outage Caused by AI-Generated Code** - Amazon's entire storefront crashed for six hours after a junior developer submitted AI-generated code, costing millions.
- 2 (03:18) **The Human Bottleneck Problem** - AI agents generate code faster than humans can review, forcing a trust trade-off that is already breaking production systems.
- 3 (04:30) **AI as Foundational Infrastructure** - AI has moved from assistive tool to bedrock of services like AWS, making outages systemic and geopolitical.
- 4 (07:38) **Lovable's Explosive Growth Proves Demand** - The no-code AI platform added $100M ARR in a single month, now at $400M, showing the trend is unstoppable.
- 5 (09:29) **Polymarket Shows OpenAI Surging Past Anthropic** - Prediction markets give OpenAI an 85% chance of having the best model by March 31, while Claude faces ongoing stability issues.
- 6 (11:04) **Anthropic's Code Review: Selling the Problem and the Solution** - Anthropic releases a paid code review product ($15-20 per review) to fix issues created by its own Claude Code, while OpenAI offers a similar tool for free.
- 7 (14:07) **Anthropic's GPU Capacity Crunch** - Claude is suffering frequent outages as Anthropic struggles to scale compute to meet surging demand, while OpenAI has more resources to subsidize costs.
+ Full timestamped outline available in the app
Guests on this episode
Show Notes
A recent six-hour Amazon outage was triggered by AI-generated code, prompting changes in their coding policies.
We explore Anthropic’s challenges with Claude Code, the rise of "Lovable," an intuitive AI app platform, and Andre Carpathy's project on self-improving AI. Join us as we navigate the evolving landscape of AI and coding.
------
🌌 LIMITLESS HQ ⬇️
NEWSLETTER: https://limitlessft.substack.com/
FOLLOW ON X: https://x.com/LimitlessFT
SPOTIFY: https://open.spotify.com/show/5oV29YUL8AzzwXkxEXlRMQ
APPLE: https://podcasts.apple.com/us/podcast/limitless-podcast/id1813210890
RSS FEED: https://limitlessft.substack.com/
------
POLYMARKET | #1 PREDICTION MARKET 🔮
https://bankless.cc/polymarket-podcast
------
TIMESTAMPS
0:0 The Amazon Outage
1:06 Growing Pains
5:27 The Job Market
7:01 Geopolitics
9:28 Rise of Lovable
10:59 Anthropic's Code Review
14:06 OpenAI vs. Anthropic
15:55 Self-Improving AI
21:08 Conundrums
------
RESOURCES
Josh: https://x.com/JoshKale
Ejaaz: https://x.com/cryptopunk7213
------
Not financial or tax advice. See our investment disclosures here:
https://www.bankless.com/disclosures
More from this podcast
Limitless Podcast →