Artificial Intelligence

3 of the 32 write-ups here are tagged Artificial Intelligence. Every figure quoted below is in the figures table with the sentence it came from.

The figures

• $1.43 — The $1.43 and the $9.05 are both frontier models doing a job they were not specifically built for. →
• $1.43 — A single run costs about $1.43 on the first and $9.05 on the second. →
• $3 — The free version replaces text on any page as often as you like, while the $3 tier adds multi-word replacement, regex matching, and rules you can set per site, which is the sort of feature split that only makes sense once you already know which users are the impatient ones. →
• $29 — Before writing a contract-comparison tool, one builder handled three to ten comparisons by hand at $29 a document, and only turned the routine into software once the same people kept coming back and paying for it. →
• 12 months — Quality degradation typically appears 6-12 months after full automation—long after the press releases have been sent. →
• $40 million — Klarna’s AI customer service experiment, which replaced 700 human agents, initially saved $40 million in a year, but the quality of service suffered so badly that they had to rehire humans, leading the CEO to admit they “went too far” with automation after 12 months. →

All figures, 335 rows

The write-ups

Debunking the Myth of Overnight Success in Micro-SaaS

A six-hour Chrome extension that pays $400 a month with no servers behind it sounds like something invented to sell a course.

The Klarna Lesson: Why AI Implementation Needs a Staircase, Not a Leap

Klarna’s AI customer service experiment, which replaced 700 human agents, initially saved $40 million in a year, but the quality of service suffered so badly that they had to rehire humans, leading…

The Two Best AI Code Reviewers Score the Same. One Costs $1.43 a Run, the Other $9.05.

On ReactBench, GPT 5.6 Sol and Fable 5 posted Pass@1 scores of 43.1% and 41.2%.

---

All 32 write-ups