Trump has posted dozens of AI-generated images and deepfake videos to his Truth Social account since returning to office in January 2025, and about half of those posts landed in August and September 2026. Most were reposts of other users' work rather than material he made himself, according to NBC News reporting on the pattern. That surge is the reason so many readers are asking the same handful of questions this month. This page answers them from documented reporting: what was actually posted, what the underlying facts were, what the law does and does not cover, and how to check a clip before you share it.
Table of Contents
- What did Trump actually post, and what was the real event?
- Is synthetic video the main source of false claims right now?
- The bigger risk is the "liar's dividend"
- Is posting a political deepfake illegal?
- How to check a clip before you share it
- Frequently Asked Questions
What did Trump actually post, and what was the real event?
The clearest case is Kharg Island. On 31 August 2026 Trump posted two AI-generated videos captioned as Iran's Kharg Island "being blown to smithereens," with no label indicating they were synthetic. Forbes reported that a US official told Reuters the military did not strike Kharg Island at all; the limited overnight strikes hit two missile launchers on Larak Island. Kharg Island is Iran's main crude oil export terminal, which is why a strike there would be enormous news. Iran's state-run National Iranian Oil Co.
chief executive Hamid Bovard publicly rejected the claim and said conditions on Kharg were "calm and appropriate," according to IranWire. Vice President JD Vance defended the post as depicting what could happen rather than what had happened. A second case was domestic. Trump posted an altered video of House Minority Leader Hakeem Jeffries wearing a sombrero and a handlebar mustache over mariachi music, hours before a government shutdown deadline. Jeffries called it "racist and fake" and told Trump not to "cop out through a racist and fake AI video," as The Hill reported; Trump then posted a further AI video of Jeffries.
Is synthetic video the main source of false claims right now?
No. The documented record of false presidential statements is still dominated by ordinary spoken claims, not fabricated footage. In his February 2026 State of the Union address, Trump said "we had the highest inflation in the history of our country," when year-over-year inflation was 2.9% in December 2024, per NBC News's fact check.
The same fact check compared his claim of 70,000 new construction jobs against the Bureau of Labor Statistics figure of 44,000 from January 2025 to January 2026. Those are verifiable numbers from a federal agency, and checking them takes a spreadsheet rather than a forensic video tool. The practical takeaway is to keep both habits. A number in a speech is usually checkable against the agency that publishes it; a video is checkable against whether any wire service or official confirms the event it depicts.
The bigger risk is the "liar's dividend"
The "liar's dividend" is a term coined by legal scholars Bobby Chesney and Danielle Citron. It describes the second-order damage from convincing fakes: once everyone knows realistic fabrication is cheap, genuine footage can be waved away as AI-generated, and the liar profits from the doubt. This is not theoretical.
During the 2026 Iran war, authentic video — including footage of Benjamin Netanyahu — was wrongly labeled AI-generated, as the concept's documented examples show. The fakes and the false accusations of faking compound each other. It also reaches well past politics. AI-generated clips of Jimmy Kimmel and Jon Stewart spread widely in August 2026, NPR reported, which means ordinary viewers now meet fabricated footage of familiar faces in normal social feeds rather than only in election contexts.
📨 Get Free Accountability Coverage Alerts
Free · No spam · Unsubscribe anytime
Is posting a political deepfake illegal?
Usually not, and readers who assume otherwise tend to overestimate their options. At least 28 US states require disclosure of, or prohibit, AI-manipulated political content near elections, according to a 2026 deepfake legislation tracker. Those statutes face recurring First Amendment challenges over parody and satire, so a deceptive political deepfake is not automatically illegal where it is posted.
The main federal law people reach for does not apply either. The TAKE IT DOWN Act's takedown obligations took effect 19 May 2026, and the FTC has sent warning letters to 15 major platforms. But the law targets nonconsensual intimate imagery — it does not give a remedy for a fabricated campaign video. What that leaves is a narrower set of tools:.
- State election statutes, which generally apply only within a defined window before an election and often only to paid or coordinated content.
- Platform policies, which are enforced privately and vary by service.
- Ordinary defamation law, which requires a false statement of fact about an identifiable person and real damages — a high bar for content framed as satire.
How to check a clip before you share it
Speed is the whole game. A fabricated video does most of its damage in the first hours, before any correction circulates, which is why the Kharg Island posts had already spread before Reuters got an official denial on the record.
- Look for a second source naming the same event. The Kharg claim collapsed because no wire service reported a strike there and an official said there was none.
- Check who the event would affect and whether they are reacting. Iran's own oil company executive said the island was calm — an unusual thing for a state firm to volunteer if the site had been hit.
- Note whether the poster labeled it. The Kharg videos carried no disclosure that they were synthetic.
- Be equally skeptical of "that's AI" claims. Real Netanyahu footage was dismissed as fake during the same conflict.
- Treat a defense of a post as evidence about the post. Vance's argument that the video showed what could happen was, in effect, confirmation that it did not show what did.
Frequently Asked Questions
Did the US strike Kharg Island in August 2026?
No. A US official told Reuters the military did not strike Kharg Island; the limited overnight strikes hit two missile launchers on Larak Island, Forbes reported.
Does the TAKE IT DOWN Act cover political deepfakes?
No. Its takedown obligations took effect 19 May 2026 but target nonconsensual intimate imagery, so it offers no remedy for a fabricated campaign video.
Did Trump make the AI videos himself?
Mostly not. NBC News found most of the AI images and deepfake videos on his Truth Social account were reposted from other users.
You Might Also Like
- Trump Science FAQ for September 2026: Source-Checked Answers to Common Questions
- Trump Medicare FAQ for September 2026: Source-Checked Answers to Common Questions
- Trump Lobbying FAQ for September 2026: Source-Checked Answers to Common Questions