Scientists in Italy hit the headlines this week, after claiming the famous Shroud of Turin dates from Jesus' lifetime around 2,000 years ago. Now, AI has reimagined what the son of God might have actually looked like based on the treasured relic, which is said to feature an imprint of Jesus' face. MailOnline asked the AI tool Merlin: 'Can you generate a realistic image of Jesus Christ based on the face in the Shroud of Turin?' The AI-generated result suggests Christ was white with big blue eyes, a trim beard and thorn marks on his face. So, can you see the similarities with the famous holy imprint? The Shroud of Turin is a 14-foot-long linen cloth with a faint image of a crucified man.

DeepFakes, which refer to AI-generated media content, have become an increasing concern due to their use as a means for disinformation. Detecting DeepFakes is currently solved with programmed machine learning algorithms. In this work, we investigate the capabilities of multimodal large language models (LLMs) in DeepFake detection. We conducted qualitative and quantitative experiments to demonstrate multimodal LLMs and show that they can expose AI-generated images through careful experimental design and prompt engineering. This is interesting, considering that LLMs are not inherently tailored for media forensic tasks, and the process does not require programming. We discuss the limitations of multimodal LLMs for these tasks and suggest possible improvements.

From text to audio and image, today's generative-AI systems are trained on large quantities of human-generated content. Most of this content is obtained by scraping a variety of online sources. As generative AI becomes more common, it is reasonable to expect that future data scraping will invariably catch generative AI's own creations. We ask what happens when these generative systems are trained on varying combinations of human-generated and AI-generated content. Although it is early in the evolution of generative AI, there is already some evidence that retraining a generative AI model on its own creation - what we call model poisoning - leads to a range of artifacts in the output of the newly trained model. It has been shown, for example, that when retrained on their own output, large language models (LLMs) contain irreversible defects that cause the model to produce gibberish - so-called model collapse [22].

It sounds like a scenario straight out of a Ridley Scott film: technology that not only sounds more "real" than actual humans, but looks more convincing, too. Yet it seems that moment has already arrived. A new study has found people are more likely to think pictures of white faces generated by AI are human than photographs of real individuals. "Remarkably, white AI faces can convincingly pass as more real than human faces – and people do not realise they are being fooled," the researchers report. The team, which includes researchers from Australia, the UK and the Netherlands, said their findings had important implications in the real world, including in identity theft, with the possibility that people could end up being duped by digital impostors.

For Taylor Klein, it starts with a weird Facebook message from a friend. Must be spam, she figures. "I think you've been hacked?" Her pal insists the link is real. She clicks, and her nightmare begins.

The exploitation of Deepfake techniques for malicious intentions has driven significant research interest in Deepfake detection. Deepfake manipulations frequently introduce random tampered traces, leading to unpredictable outcomes in different facial regions. However, existing detection methods heavily rely on specific forgery indicators, and as the forgery mode improves, these traces become increasingly randomized, resulting in a decline in the detection performance of methods reliant on specific forgery traces. To address the limitation, we propose Recap, a novel Deepfake detection model that exposes unspecific facial part inconsistencies by recovering faces and enlarges the differences between real and fake by mapping recovered faces. In the recovering stage, the model focuses on randomly masking regions of interest (ROIs) and reconstructing real faces without unpredictable tampered traces, resulting in a relatively good recovery effect for real faces while a poor recovery effect for fake faces. In the mapping stage, the output of the recovery phase serves as supervision to guide the facial mapping process. This mapping process strategically emphasizes the mapping of fake faces with poor recovery, leading to a further deterioration in their representation, while enhancing and refining the mapping of real faces with good representation. As a result, this approach significantly amplifies the discrepancies between real and fake videos. Our extensive experiments on standard benchmarks demonstrate that Recap is effective in multiple scenarios.

Adversarial attacks aim to disturb the functionality of a target system by adding specific noise to the input samples, bringing potential threats to security and robustness when applied to facial recognition systems. Although existing defense techniques achieve high accuracy in detecting some specific adversarial faces (adv-faces), new attack methods especially GAN-based attacks with completely different noise patterns circumvent them and reach a higher attack success rate. Even worse, existing techniques require attack data before implementing the defense, making it impractical to defend newly emerging attacks that are unseen to defenders. In this paper, we investigate the intrinsic generality of adv-faces and propose to generate pseudo adv-faces by perturbing real faces with three heuristically designed noise patterns. We are the first to train an adv-face detector using only real faces and their self-perturbations, agnostic to victim facial recognition systems, and agnostic to unseen attacks. By regarding adv-faces as out-of-distribution data, we then naturally introduce a novel cascaded system for adv-face detection, which consists of training data self-perturbations, decision boundary regularization, and a max-pooling-based binary classifier focusing on abnormal local color aberrations. Experiments conducted on LFW and CelebA-HQ datasets with eight gradient-based and two GAN-based attacks validate that our method generalizes to a variety of unseen adversarial attacks.

The growing reliance of society on social media for authentic information has done nothing but increase over the past years. This has only raised the potential consequences of the spread of misinformation. One of the growing methods in popularity is to deceive users using a deepfake. A deepfake is an invention that has come with the latest technological advancements, which enables nefarious online users to replace their face with a computer generated, synthetic face of numerous powerful members of society. Deepfake images and videos now provide the means to mimic important political and cultural figures to spread massive amounts of false information. Models that can detect these deepfakes to prevent the spread of misinformation are now of tremendous necessity. In this paper, we propose a new deepfake detection schema utilizing two deep learning algorithms: long short term memory and multilayer perceptron. We evaluate our model using a publicly available dataset named 140k Real and Fake Faces to detect images altered by a deepfake with accuracies achieved as high as 74.7%

Abstract: The emergence of deepfake technologies has become a matter of social concern as they pose threats to individual privacy and public security. It is now of great significance to develop reliable deepfake detectors. However, with numerous face manipulation algorithms present, it is almost impossible to collect sufficient representative fake faces, and it is hard for existing detectors to generalize to all types of manipulation. Therefore, we turn to learn the distribution of real faces, and indirectly identify fake images that deviate from the real face distribution. In this study, we propose Real Face Foundation Representation Learning (RFFR), which aims to learn a general representation from large-scale real face datasets and detect potential artifacts outside the distribution of RFFR. Specifically, we train a model on real face datasets by masked image modeling (MIM), which results in a discrepancy between input faces and the reconstructed ones when applying the model on fake samples.

We are excited to bring Transform 2022 back in-person July 19 and virtually July 20 - 28. Join AI and data leaders for insightful talks and exciting networking opportunities. Earlier this year a chilling academic study was published by researchers at Lancaster University and UC Berkeley. Using a sophisticated form of AI known as a GAN (Generative Adversarial Network) they created artificial human faces (i.e. They discovered that this type of AI technology has become so effective, we humans can no longer tell the difference between real people and virtual people (or "veeple" as I call them). You see, they also asked their test subjects to rate the "trustworthiness" of each face and discovered that consumers find AI-generated faces to be significantly more trustworthy than real faces.