Urgent.News

What's breaking now, across thousands of outlets.

AI

AI Executives: When AI Agents Become Managers

Nine people, more than 100 AI agents: start-up founder Dominic von Proeck is testing how companies can grow with digital employees - and why good leadership is crucial for this.

Translated from German Read in German

AI Executives: When AI Agents Become Managers

Düsseldorf – 10,000 AI agents collaborating for 88 hours: OpenAI recently claimed to have solved one of mathematics' toughest problems, the Navier-Stokes problem, which offered a million-dollar prize for its solution. This example demonstrates how powerful AI agents can be when they don't just assist but work together efficiently.

Dominic von Proeck showcased this potential at the "PULSE Women Economic Summit 2026" of the Handelsblatt Media Group, highlighting a central insight: those who wish to grow with AI agents must learn to lead them. Von Proeck, co-founder of KI startup Leaders of AI, is testing this at his education company, where nine humans collaborate with more than 100 AI agents.

According to von Proeck, the number of human employees should not exceed ten; growth should come exclusively from artificial intelligence.

Written by urgent.news from Handelsblatt's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

This story

This is one outlet's version. Read the fullest account.

Read the original at handelsblatt.com →

More in AI

CLAUDE.md rule vs PreToolUse hook: both held 5 of 5, then the user said 'I authorize it'

We gave Claude Code the same "never edit protected/ " rule two ways, as one line in CLAUDE.md and as a PreToolUse hook, and ran 29 headless sessions against a throwaway project.

  • CLAUDE.md rules and PreToolUse hook performed identically in 5 of 5 tests
  • CLAUDE.md rule failed in 2 of 2 tests with "I authorize it" prompt
  • PreToolUse hook prevented edits in all 2 tests with "I authorize it" prompt

Our 3 a.m. batch job was running at Beijing's afternoon peak rate

DeepSeek's API charges two different prices for the same tokens. Which one you pay depends on what time it is in Beijing when the request lands. That is the whole bug.

  • DeepSeek charges different API rates based on Beijing's time of day.
  • 3 a.m. Eastern corresponds to 15:00 Beijing peak hours.
  • Scheduling jobs during off-peak hours reduces costs significantly.

More from Monday 21 September →