The BBC Ran a Secret AI Marketing Test on Doctor Who Fans, and They Noticed
What happened
In March 2024, the BBC sent promotional emails for its Doctor Who series that had been produced using generative AI. Fans noticed immediately. The corporation received a flood of complaints and within days announced it would not use the technology again to promote the show.
The emails were part of what the BBC described as a small trial. Davis Housden, the BBC's Head of Media Inventory, had articulated the logic behind it plainly: generative AI offers an opportunity to speed up the production of extra assets and get more promotional experiments running across more content. That framing treats creative work as a throughput problem, where the bottleneck is the time a human writer takes to produce copy, and the AI is the tool that removes it. It is a reasonable efficiency argument, but it was made internally and the audience receiving the output was not told any of this.
The backlash came from two directions at once. Doctor Who fans objected to what the AI-generated promotions felt like, finding them at odds with a show they had a long and invested relationship with. Media professionals, including writers, marketing staff, and PR practitioners, used the moment to say something broader: that trials like this are not experiments in efficiency so much as announcements about which jobs are scheduled for replacement. The concern was not abstract. The BBC's own creative and editorial workforce sits in exactly the category those professionals were describing.
The BBC moved quickly to end the experiment. Its statement confirmed the AI trial for Doctor Who promotions was terminated and would not be repeated for that property. That resolved the immediate public complaint, but it left several questions untouched. The BBC did not say whether other shows were part of similar trials, whether additional AI-generated promotional material had already been distributed, or what approval process existed for using AI in audience-facing content at a public broadcaster.
That last gap is the structural problem. The trial ran, generated a public backlash, and was shut down, but nothing in the public record shows who authorized the use of AI for content going directly to the BBC's audience, what the system produced before any human reviewed it, or how the decision to run the trial was made and documented. Without a provable record of what a system did and who approved each step, any institution can describe a deployment as a small trial and discontinue it under pressure, with no accounting for what reached the public or why.
Reported impact
- Affected parties
- Not publicly disclosed
- Harm type
- Not publicly disclosed
- Scale
- Not publicly disclosed
- Financial impact
- Not publicly disclosed
- Regulatory action
- Not publicly disclosed
Classification
Relevant governance controls
Governance control mapping is not available for this record.
- No controls mapped
Not publicly disclosed
Control mapping is analytical. It does not state that any control would have prevented the incident.
Sources and evidence
This record was researched and written by the Index. The event is also catalogued in the following database, which is listed for cross-reference.