More teachers than ever are trained on AI. Are they ready to use it?

The number of districts that have trained teachers in generative artificial intelligence have more than doubled from 2025 to 2025, according to a new analysis of the research company and Politics Rand Corp.

In the fall of 2025, 48% of the districts said they had trained teachers to use AI, against 23% in the fall of 2025. Rand questioned around 300 districts in 2025 and around 200 in 2025 and weighed their responses to make them representative of all public school districts.

If the districts follow their plans for the current school year, almost three -quarters will have given training on AI to educators by fall 2025.

This professional development was slow to gain ground in schools. While generative AI became the general public at the end of 2022 with the launch of Chatgpt, 7 in 10 teachers had not received any training on how to use AI in class In the spring of 2025, according to a representative survey at the national level of the Edweek Research Center. In the fall of 2025, however, 43% of teachers said they had received at least one training session on AIThe Edweek Research Center has found.

Teachers and directors are increasingly using AI in their daily work. While directors use AI tools for administrative tasks Like creating schedules, writing newsletters or cracking large sets of data, teachers are more likely to use AI to generate lessons, personalize lessons for students who could learn at different speeds and levels, or even use in a creative way – to imitate a historical Or a fictitious character to teach a particular lesson, for example.

Despite this, most districts have adopted a progressive approach to inform their teachers of the AI.

In addition to setting up surveys, Rand has conducted in -depth interviews with 14 district chiefs on how they designed their IA training in 2025. Instead of jumping directly in AI tools and their use, district chiefs first focused on the fight against teachers concerning technology.

Of the 14 leaders interviewed, all except one declared that they had encountered “negative opinions of AI and reported that some teachers considered it a threat to traditional teaching methods or as a student cheating tool”, according to the report.

The district chiefs said they had provided training on the fundamental functioning of AI – and its limits – to reduce teachers' anxiety and “move away from an antagonistic mentality and centered on cheating” on AI.

Among the 14 districts, the training varied in length and content, from one day training to the content “the size of a bite” delivered by regular emails and newsletters. All districts except one has made such voluntary training.

“I am not trying to push AI with my people, but I try to give them the tools if they are interested. I know that (the adoption of AI) begins to happen, it will develop in a organic way and very quickly,” a district chief for researchers told Rand.

District chiefs tend to create their own training

The district chiefs interviewed by Rand preferred to create their own training courses, rather than relying on external experts, said Melissa Diliberti, associate researcher with Rand and the report of the report, in an email.

“This DIY approach, although potentially used to meet the specific needs of the district, also reflects a scarcity of external experts capable of providing appropriate training,” said Diliberti.

Of the 14 district chiefs questioned, 11 adopted a DIY approach to develop their own training on AI. They have reconstructed different types of resources published by technological companies such as Google, as well as educational organizations as a promised digital and the international society for education technology.

Half of the respondents said they had trouble finding training experts who knew the use of AI in an educational context.

A time and resources crisis has also prompted district chiefs to bring together their own training, in the absence of ready -to -use resources, according to the report. Professional development in IA lullabies for space among several other competing PD priorities, which can affect the duration and depth of such training, said district chiefs.

District chiefs have also used their own “negative” experiences with AI training – where they were exceeded – to rationalize the training they have offered so that it meets the immediate needs of teachers.

A district of the west coast of the suburbs, for example, first introduced “game” sessions to alleviate teachers' anxiety about AI, before gradually moving to learning more based on teaching.

The “game” approach, in this district and others, consisted of teachers experimenting with a chatbot or an AI platform and to understand what AI can do. There have also been more targeted sessions in which district chiefs have introduced tools like magicschool or chatgpt to show teachers how to use them to plan lessons or generate differentiated content for students.

Chris Chism, the Superintendent of the Mississippi Pearl School District, who was not associated with the Rand report, also created his own AI training module for teachers. His training also highlights both experimenting with technology and presenting more rationalized and direct uses for the classroom.

In an interview with education week last July, Chism said he had created Mini chatbots which provide essential information to teachers on specific subjects. A chatbot, for example, can generate course plans aligned with state standards.

While the district leaders' training can be more in line with the needs of a teacher, Diliberti said that quality and what is covered in these training courses could vary between districts.

There is a equity gap in IA training

Despite the global overvoltage of AI training, the Rand report shows that such a training is unevenly distributed – carrying out almost 67% of low -poverty districts that introduced training on the teachers, only 39% of high pause districts were able to do the same.

Rand used the district plans to project what the training gap in the fall of 2025 might look like almost the entire district with low poverty would have formed teachers in the use of AI, only 6 districts on high pauses would have made, if the districts stand in their plans.

Schools of low poverty districts are more likely to have time and resources to experiment with AI tools and find how to use them well for educational practice. It is not a luxury that schools in high pause districts can have, said Diliberti.

This gap in training or advice, on how to use AI, also persists among the main group Previous hiking report Posted in February. Thirteen percent of the highest school directors said they had obtained advice, compared to approximately a quarter of directors in low -break schools.

“Teachers in higher poverty schools are less likely to use AI, (as are) directors in the highest schools.

The faster takeover of AI by teachers in low poverty districts means that teachers and high -poverty students will not have the same possibility of learning AI tools, according to the report.

There are still no clear ties between greater dependence on AI tools and better quality teaching or learning, Diliberti said.

“But insofar as we plan that this could be true, greater absorption of AI in historically advantageous school circles raises fear that the deployment of AI in schools can exacerbate the long -standing flaw lines from educational inequalities,” added Diliberti.

The report requires more targeted funding by federal and state education agencies to high pause districts in order to obtain a professional development model in AI and cultivate links with expert networks.

The report also suggested that districts with more experience in AI teachers' training should share their expertise with other districts through national and regional education networks.

Despite the sharp increase in training, Diliberti said it was still too early to understand how AI could adapt to a District's MP calendar for the year.

“The evidence of our district chiefs suggest that their training is in the early stages. (They are) more focused on putting teachers at ease with AI tools,” said Diliberti. “Regular training of teachers on the question of whether and how to best integrate specific AI tools into pedagogical practice can still be absence.”