“Nous avons demandé à trois Intelligences Artificielles (DALL.E, MIDJOUNEY & STABLE DIFFUSION) de générer des images. Laurence DEVILLER, Professeure en informatique et IA à la Sorbonne Université, et Albertine MEUNIER, Artiste Numérique et Chevalier de la Légion, d’Honneur-sauront elles les différencier d’oeuvres humaines ?”
Auteur/autrice : noflux (Page 27 of 614)
“San Francisco’s police will be allowed to use remote-controlled robots to kill suspects. The city’s board of supervisors last night approved a controversial policy that lets police robots “be used as a deadly force option when risk of loss of life to members of the public or officers is imminent and outweighs any other force option available.”
The San Francisco Police Department (SFPD) said it does not own any pre-armed robots and has no plans to arm its current machines, reports Sky News. As SFPD spokesperson Allison Maxie explained in a statement, the department’s robots can now be equipped with explosives “to contact, incapacitate, or disorient violent, armed, or dangerous suspect” in “extreme circumstances to save or prevent further loss of innocent lives.””
Source : San Francisco approves use of remote-controlled robots to kill suspects – The Verge
“DALL·E’s Edit feature already enables changes within a generated or uploaded image — a capability known as Inpainting. Now, with Outpainting, users can extend the original image, creating large-scale images in any aspect ratio. Outpainting takes into account the image’s existing visual elements — including shadows, reflections, and textures — to maintain the context of the original image.”
Source : DALL·E: Introducing Outpainting
“We’ve trained a model called ChatGPT which interacts in a conversational way. The dialogue format makes it possible for ChatGPT to answer followup questions, admit its mistakes, challenge incorrect premises, and reject inappropriate requests. ChatGPT is a sibling model to InstructGPT, which is trained to follow an instruction in a prompt and provide a detailed response.
We are excited to introduce ChatGPT to get users’ feedback and learn about its strengths and weaknesses. During the research preview, usage of ChatGPT is free.”
Try it now at chat.openai.com.
“As part of Intel’s Responsible AI work, the company has productized FakeCatcher, a technology that can detect fake videos with a 96% accuracy rate. Intel’s deepfake detection platform is the world’s first real-time deepfake detector that returns results in milliseconds.
Most deep learning-based detectors look at raw data to try to find signs of inauthenticity and identify what is wrong with a video. In contrast, FakeCatcher looks for authentic clues in real videos, by assessing what makes us human— subtle “blood flow” in the pixels of a video. When our hearts pump blood, our veins change color. These blood flow signals are collected from all over the face and algorithms translate these signals into spatiotemporal maps. Then, using deep learning, we can instantly detect whether a video is real or fake. ”
“Edmondson has no plans to make the device into a commercial product, but he says the design could easily be copied and reused by anyone with some technical knowledge. Many of the parts involved are easy to obtain or may be lying around the homes of people in tech communities.
Ultimately, he says, the tech community needs to take tech-enabled tracking and surveillance more seriously. “It was really kind of disheartening and depressing to look at the ratio of tools to spy on people versus tools to help you not get spied on,” he says, adding that a person close to him has been the victim of a stalker in the past. In the case of his clandestine friend in another government department, the anti-tracking device was useful. “It was really designed to help someone who came to me asking for help,” he says. Fortunately for Edmondson’s friend (and his source), they used it in the real world, and the device didn’t find anyone following them.”
Source : This Anti-Tracking Tool Checks If You’re Being Followed | WIRED
“Alors que le monde d’aujourd’hui est confronté aux risques de rupture climatique et de pénuries énergétiques, ce livre ose imaginer comment d’innombrables actes de partage, libres, justes, et vivants, peuvent construire une nouvelle culture et une économie politique révolutionnaire. Il se donne un objectif ambitieux : celui de conceptualiser les communs en tant que système social.”
“Facebook’s decision to get rid of these specific profile fields is part of its efforts to streamline its platform, which currently consists of several features that are somewhat outdated. It’s worth noting that the information fields that Facebook is choosing to remove are ones that other major social networks don’t offer. Platforms like Instagram and TikTok have simple bios that let users share a little bit about themselves without going into specific details, such as political or religious views. In the past, people may have been interested in filling out their profiles with additional information, but as privacy infringements have come to light, users may no longer want to share extra details about themselves online.”
“«Jamais dans ma carrière je n’ai vu une telle absence de contrôles internes et d’informations financières fiables», a résumé le liquidateur à propos de FTX (qui s’est également fait hacker 400 millions de dollars la veille de sa faillite), déplorant au passage que le pouvoir de contrôle ait été «concentré dans les mains d’un très petit nombre d’individus sans expérience, peu sophistiqués et potentiellement compromis». Une situation sans précédent, pour le liquidateur, qui sait de quoi il parle: il s’était occupé de la faillite d’Enron, l’une des plus grandes et plus complexes fraudes de l’histoire moderne, la mère de toutes les manipulations comptables.”
“Depuis lundi, les quelque 50 000 entreprises qui utilisent le logiciel de facturation Winbiz n’ont plus accès aux donnés de leur cloud. La société créée à Martigny et rachetée l’an dernier par le géant français Fiducial explique que son hébergeur de données infopro.ch a subi une attaque informatique: «Par mesure de précaution et afin d’évaluer l’impact de manière précise, l’ensemble des serveurs de celui-ci ont été arrêtés, y compris les nôtres, afin qu’ils puissent conduire une analyse détaillée et éviter tout risque de propagation.»”
Source : Une cyberattaque paralyse le paiement des salaires de 50 000 entreprises – Le Temps