The widely shared image of Senator Robin Padilla reading a book about Albert Einstein was an AI-manipulated edit of an authentic August 27 photo showing him reviewing the proposed 2027 national budget. Fact-checkers identified telltale errors — notably the misspellings "Eistein" and "pampataling" — and OpenAI's verification tool detected an invisible "SynthID" watermark, indicating AI origin. Reverse-image searches located the original budget photo, and AFP debunked the doctored image.
AI-Altered 'Einstein' Image Mocking Senator Robin Padilla Traced To Budget Photo; Invisible SynthID Detected
Screenshot of the false post, taken on September 22, 2026, with a red X and an AI label added by AFP
An AI-manipulated image mocking Philippine Senator Robin Padilla circulated widely on social media after his remarks during the impeachment trial of Vice President Sara Duterte. The doctored photo — which replaces the original subject matter with a book about Albert Einstein and other physics props — was edited from an authentic photo of Padilla reviewing the proposed 2027 national budget.
Screenshot of the AI-manipulated image, visual errors highlighted by AFP
What Happened
On August 26 senators acting as the impeachment court reviewed documents related to alleged confidential funds from the vice president's office. Padilla drew attention during the proceedings for defending the use of aliases, citing historical examples. His comments prompted criticism, including a rebuttal from Senator Paolo Benigno "Bam" Aquino over a comparison involving his late uncle, Benigno "Ninoy" Aquino.
Screenshot comparison of the AI-manipulated image (L) and the original photo uploaded to Senator Robin Padilla's Facebook page
How The Image Was Manipulated
Fact-checkers identified multiple visual errors in the viral image that are typical of AI generation. Observers noted misspellings on visible props — most prominently the book title rendered as "Eistein" and a drink label reading "pampataling" (a mistaken rendering of the Tagalog word "pampatalino"). Those anomalies helped trigger additional verification checks.
Screenshot of the OpenAI verification site's analysis
Key Technical Finding: A reverse-image search located the original August 27 Facebook post from Senator Padilla showing him reading President Ferdinand Marcos's message on the proposed 2027 national budget. An analysis by OpenAI's verification tool also found a hidden "SynthID" watermark in the altered image, indicating it originated from an AI image-generation system and that the watermark survived subsequent edits.
Social Reaction
The manipulated image circulated alongside mocking Tagalog captions and comments such as "Don't be pretentious. Go back to reading about the ABCs and colours," and disparaging remarks about Padilla's grasp of English and law. AFP flagged the post, added an AI label to a screenshot, and published verification results debunking the image.
Why It Matters
This episode underscores how easily AI tools can produce convincing yet flawed political imagery and how those images can shape public perception during sensitive proceedings. Visible errors (misspellings and odd object details), reverse-image tracing to the original budget photo, and detection of an invisible SynthID watermark together provide a clear chain of evidence that the widely shared photo was fabricated.
Bottom line: The viral Einstein photo is not genuine. It was AI-altered from a genuine budget-related photo of Senator Padilla, and both visual forensics and digital watermarking point to AI generation.
Help us improve.
