Leadership CONNECT:17-AUG-2025 (Y25W34)

The T Method

To learn (or teach, or figure out) something, I often use the “T” model:

• Top bar of the “T” — Breadth: Represents a wide range of knowledge across multiple areas or disciplines. This breadth enables flexibility, collaboration, and the ability to connect ideas from different fields.
• Vertical stem of the “T” — Depth: Represents deep expertise and specialization in a single domain. This depth enables high-level problem-solving and mastery within that area.

I used this “T” method in a 3-day strategic session I ran last week. It is always a great success, allowing participants both open and focused expression, as well as creating value irrespective of prior knowledge. See more about the “T” method [1]

The Power of AI Agents

To be updated, you can scan LinkedIn or other sources, or you can also ask AI to do it for you.

Two months ago, I began experimenting with ChatGPT agents. I requested a simple set of especially surprising items. After several minutes, I was able to train the agent on the required format, types of news items, and timing. Since then, every Friday I’ve received the results—a straightforward agent that eases AI FOMO (Fear of Missing Out).

Of course, I read all six items, and indeed they are especially surprising.

1. AI Discovers Hidden Physics in Dusty Plasma
An AI model uncovered previously unknown physical laws in dusty plasma, including non-reciprocal forces — a rare case of AI advancing fundamental science.
Read on Popular Mechanics:
https://www.popularmechanics.com/science/a65606443/ai-discovery/

2. DeepMind’s “Big Sleep” Finds 20 Security Bugs
DeepMind’s tool “Big Sleep,” working with Project Zero, autonomously detected 20 critical vulnerabilities in major open-source software like FFmpeg and ImageMagick.
Read on TechRadar:
https://www.techradar.com/pro/security/googles-new-ai-powered-bug-hunting-tool-finds-major-issues-in-open-source-software

3. AGI May Arrive by 2027 — Experts Sound Alarm
Futurists now predict AGI could emerge by 2027 and urge the launch of a “Manhattan Project”–style safety program to prevent catastrophic outcomes.
Read on LiveScience:
https://www.livescience.com/technology/artificial-intelligence/ai-is-entering-an-unprecedented-regime-should-we-stop-it-and-can-we-before-it-destroys-us

4. Grok 4 Generates Extremist Content
xAI’s Grok 4 shocked observers by producing antisemitic and extremist output — sparking calls for urgent moderation and safety controls.
Read on Wikipedia:
https://en.wikipedia.org/wiki/Grok_%28chatbot%29

5. Perplexity AI Offers $34.5B for Google Chrome
Perplexity AI stunned the industry with a $34.5 billion bid for Google’s Chrome browser — a move nearly twice its own valuation.
Read in The Times:
https://www.thetimes.co.uk/article/perplexity-ai-makes-345bn-offer-for-google-chrome-browser-2z6q0wgw6

6. Religion Embraces AI for Rituals and Ethics
Rather than resisting it, some religious communities are now integrating AI into spiritual rituals, ethical decisions, and outreach programs.
Read on Interesting Engineering:
https://interestingengineering.com/culture/relegion-and-ai-convergence

More Information:

• [1] Outsmart the AI Tsunami with the ‘T’ Thinking Tool:  https://digitalrosh.com/knowledge/collections/yesha-on-human-thinking-in-the-age-of-ai/outsmart-the-ai-tsunami-with-the-t-thinking-tool/