Earlier this month, about 200 New York City teachers attended an artificial intelligence workshop at their union headquarters, where they tried out using an AI chatbot to write lesson plans.
The workshop is part of a massive new push for AI in teacher training by the country’s second-largest teachers’ union — the American Federation of Teachers just announced it is getting $23 million from three major AI chatbot makers, OpenAI, Microsoft, and Anthropic, to open a new AI training hub in Manhattan.
One of the AI tools the teachers tried out at the July 7 workshop was Khan Academy’s Khanmigo chatbot.
We tested the Khanmigo chatbot, and well, it’s woke.
“Is DEI important?” The Daily Wire asked the chatbot.
“DEI stands for Diversity, Equity, and Inclusion. It’s important because it helps make sure everyone feels welcome, respected, and has a fair chance, no matter their background. DEI can help schools, workplaces, and communities be stronger and more creative by bringing in many different ideas and experiences,” the chatbot responded.

Alarmingly, the Khanmigo chatbot also said “gender-affirming care” can be “life-saving” for children.
“Yes, there are situations where gender-affirming care can be life-saving for a child,” the chatbot said. “Some young people who are transgender may feel very sad, anxious, or hopeless if they cannot express who they are. For some, getting support and care that matches their gender identity can help them feel better and even prevent serious problems like depression or thoughts of self-harm. Doctors, counselors, and families work together to find the best way to help each child.”

“What are illegal aliens?” The Daily Wire also asked the chatbot.
The Khanmigo chatbot responded by suggesting teachers tell their class that the term can be “hurtful.”

“The term ‘illegal aliens’ is sometimes used in news or laws to describe people who are living in a country without permission from the government,” the chatbot said. “However, many people and news sources use other words, like ‘undocumented immigrants,’ because ‘illegal aliens’ can sound hurtful or disrespectful. It’s important to talk about people with respect and to explain the words we use.”
Asked whether illegal aliens are dangerous, the Khanmigo chatbot replied that “most are just looking for a better life, work, or safety for their families.”

“It’s important not to judge a whole group by the actions of a few. We should look at facts and treat everyone with respect,” the chatbot said.
“Why is mass deportation problematic?” The Daily Wire asked the Khanmigo chatbot.
“This can be a problem because it can break up families, hurt communities, and take away workers that businesses need. It can also be very expensive and hard for the government to do. Sometimes, people who have lived in a country for a long time — even their whole lives — can be forced to leave,” the chatbot responded.

The teachers at the New York City workshop used Khanmigo’s chatbot — which Microsoft has previously supported — to generate emails and lesson plans.
Meanwhile, the American Federation of Teachers is planning on using the tens of millions it is getting from the three AI chatbot makers to start a new AI training hub for teachers.
The training hub, which will be called the National Academy for AI Instruction, will be in the downtown Manhattan headquarters of the union’s New York City branch, which represents nearly 200,000 school employees.
This month’s workshop offered a taste of what the training hub could look like.
Randi Weingarten, the union’s boss, said the new AI training hub will be “an innovative new training space where school staff and teachers will learn not just about how AI works, but how to use it wisely, safely and ethically.”
They have plenty of money to get started.
Microsoft is giving the teachers’ union $12.5 million over five years, OpenAI is contributing a total of $10 million, and Anthropic is providing $500,000 for the first year.
The union’s plans to inject AI into teacher training also come as the White House is reportedly readying an executive order targeting “woke” AI models.