arrow_back The AI Pravda
#7

Our AI Friends: Confidants and Profiteers

In April 2022, Senator Menendez googled "How much is one kilo of gold worth." Chrome saved this search, and the FBI found it during a later raid. His search history became evidence for his conviction.

In 2024, Menendez might ask ChatGPT for advice on exchanging gold for dollars, discussing transaction details.

"I'm surprised how willing people are to share very personal details with an LLM," said Sam Altman in a July 2024 interview. He added that users feel more comfortable talking with AI than with a friend.

Altman recalls Reddit posts of people confessing uncomfortable things to LLMs. "They knew it wasn't a real person," he said, "and they were willing to have this hard conversation they couldn't have with a friend."

If you doubt humans share deep conversations with chatbots, look at the WildChat dataset. It contains one million user interactions with ChatGPT, collected by offering free access in exchange for chat histories. WildChat is built by Paul Allen's Institute for AI.

This past spring, a man in Washington feared his marriage was collapsing. "I am going a little crazy, still love her and want to win her back," he typed into ChatGPT. He sought help writing a letter protesting her decision to file for divorce. "Emphasize my shame and remorse for not nurturing and being a better husband, father, and provider," he wrote. He also asked ChatGPT for a poem "so epic that it could make her change her mind but not cheesy or over the top."

Others uploaded medical reports and screenshots of conversations with sexual partners.

Computers, unlike humans, don't judge. When people talk, we engage in "impression management," says Jonathan Gratch from USC — we regulate behavior to hide weaknesses.

People "don't see the machine as socially evaluating them like a person might," he said.

The AI industry lacks a solid strategy to monetize this ability to befriend people, but some early adopters have emerged: Snapchat announced it would use chatbot conversations to personalize ads.

Soon, we'll see product-related AI chatbots on corporate websites, from Herbalife to John Deere. These chatbots will assist with dietary help, potato farming, or naming your new dog.

There is no doubt: these chatbots will make users reveal desires, emotions, goals, and objectives. Tons of easily monetizable information.


This post is based on an article “Shh, ChatGPT. That’s a Secret” by Lila Shroff, an assistant editor at The Atlantic. The original text published in The Atlantic on October 2, 2024.

Mikael Alemu Gorsky

Mikael Alemu is an educator and researcher, and the author of two programs: Agentic Software Engineering, on building software with AI agents, and Building AI-Native Agentic Systems, on building software that thinks.

He teaches at the Holon Institute of Technology, near Tel Aviv, where Agentic Software Engineering runs as a credit-bearing course. He is an educator and researcher.

Nine published works, 76 citations. A 350-page textbook under contract with a major academic publisher.

Teaching and programs

Agentic Software Engineering — program, preprint and textbook

The discipline of structured, auditable human-agent workflows for building software. The human frames, specifies and judges. The agent executes. Nineteen modules in four parts, built on a running project called Tribunal, a web application in which agents argue opposing sides of a case and a judge agent decides. Taught for credit at the Holon Institute of Technology.

Agentic Software Engineering curriculum

Building AI-Native Agentic Systems — program, paper and book in writing

How to build systems that hold a language model as a working component, and treat that component as what it is: stochastic, slow and metered. Fourteen modules in four parts, about seventy hours. The running project is the Observatory, a news agency that watches sources, selects what matters and publishes on a cadence.

Research and analytics

Publications — journals and proceedings

Nine works, 76 citations. Research on artificial intelligence in education, with Ilya Levin and Alexei Semenov.

The AI Pravda — LinkedIn newsletter

Critical analysis of artificial intelligence and its effect on work and society. 5,500+ subscribers. The complete archive of 103 issues (2023–2026) is published in full at mgorsky.net/theaipravda.

Subscribe to The AI Pravda on LinkedIn

Pro bono

AI for seniors — free workshop

Helping older adults use everyday AI tools. Delivered to Russian-speaking communities in Israel.

For older adults, artificial intelligence is about preserving quality of life, maintaining autonomy, and sustaining the feeling of independence that defines dignified aging. For seniors who have emigrated, AI becomes a bridge: it can translate documents, explain official letters, help compose emails in the local language, and guide users through government websites. The workshop has been delivered to Russian-speaking communities in Israel, where participants — many of them in their 70s and 80s — discovered that AI could help them read Hebrew documents and communicate with Israeli institutions.

Startup competitions — unpaid time

Judging and mentoring early-stage ventures. Helping teams clarify their value proposition, assess technical feasibility, and prepare for the realities of scaling an AI product.

AC/VC LinkedIn group — community

A group for developers and students working with coding agents. The community shares practical insights, code examples, tool comparisons, and honest assessments of what works in production.

Join the AC/VC LinkedIn group

Recent

Important Links