PulseAugur
EN
LIVE 09:53:45

Multimodal LLMs show promise in detecting AI-generated deepfakes

By PulseAugur Editorial · [1 sources] ·

A new research paper explores the potential of multimodal large language models (MLLMs) to detect AI-generated deepfake facial images. The study benchmarked several MLLMs, including GPT-4o, GPT-5, GPT-5.2, Gemini 2.5, and Seed 1.6, using a dataset of authentic and generated images. Results showed high detection accuracy, with GPT-5.2 achieving 97.9% in pairwise comparisons by identifying structural inconsistencies. However, the presence of SynthID watermarks reduced Gemini's detection accuracy by 10%, suggesting that while MLLMs can detect artifacts without watermarks, their performance is affected by such embeddings. AI

IMPACT MLLMs show potential for deepfake detection, but performance can be impacted by watermarking techniques.

RANK_REASON Research paper published on arXiv detailing the use of MLLMs for deepfake detection. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Multimodal LLMs show promise in detecting AI-generated deepfakes

How we ranked this

Signal score
12 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
Research paper published on arXiv detailing the use of MLLMs for deepfake detection. [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
paper, safety
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. arXiv cs.CV TIER_1 English(EN) · Olivia Salamon, Virginia N. L. Franqueira, Shujun Li ·

    Beyond Generation: Using Multimodal LLMs as Detectors of LLM-Generated Deepfake Facial Images

    arXiv:2610.21760v1 Announce Type: new Abstract: As image generation models become increasingly capable of producing highly realistic content, traditional forensic techniques are struggling to keep up. This paper investigates whether MLLMs can instead be used as a form of forensic…

RELATED ENTITIES

RELATED TOPICS