Full Breakdown
AI Models Rate Former President Donald Trump's IQ as Low-Average
By Drooid · · How we work
Core Event: AI-Driven IQ Estimates
A YouTuber known as David posted a video in which he fed transcripts of former President Donald Trump's recent interviews to seven generative-AI models and asked each to assign an IQ score. The models evaluated the language in the transcripts and produced a range of estimates, most placing Trump in the low-average to average range.
Background & Context
Donald Trump has repeatedly asserted high intelligence. In 2018 he posted on Twitter that he was “a very stable genius.” Those statements prompted the YouTuber’s experiment.
Data & Statistics: Model Estimates
- Grok (xAI) – estimated a range of 95 to 110, describing the speech as “low average to average.”
- Meta AI (Meta) – gave a “low average” estimate of roughly 85 to 95, noting simple, repetitive vocabulary.
- DeepSeek – assigned an IQ of about 80 to 85, labeling the reasoning “low average at best.”
- Google Gemini (Google DeepMind) – placed Trump in the 85 to 95 range, citing elementary vocabulary and fragmented thoughts.
- Truth Social AI – rated him “average to somewhat below average,” roughly 95 to 105.
- ChatGPT (OpenAI) and Microsoft Copilot – declined to provide a numeric score, stating the transcripts lacked a defensible basis, but both described his verbal performance as “highly uneven” and noted occasional “sharp tactical thinking.”
Official Statements & Responses
At the October 3 rally, Trump claimed his intelligence was “very high” and that his IQ ranked among the world’s highest. No official response from the AI developers was recorded beyond the model outputs described above.
Conflicting Reports & Gaps
The AI assessments vary widely. Four models supplied numeric ranges between 80 and 110, while two models refused to assign a score, citing insufficient data. The methodology each model used to translate transcript analysis into an IQ estimate was not disclosed, leaving uncertainty about the reliability of the figures.
Why It Matters
The experiment illustrates how generative-AI systems interpret public speech and highlights the limitations of using AI to approximate standardized intelligence measures. The divergent outputs also underscore the lack of consensus among AI models on evaluating verbal reasoning from limited textual data.
Timeline
- October 3 – Trump makes high-IQ claim at Ohio rally.
- Subsequent days – David releases the video showing AI model estimates (exact release date not specified in sources).