AI image generation

A photograph from India is circulating widely online with the caption 'real photo, not AI', stressing its authenticity. The image is drawing large engagement as viewers debate what it shows and why the poster felt the need to clarify it was not artificially generated.

A photographer on Mastodon has posted an urban street photograph under the photo suggestion 'in the city' (FotoVorschlag), noting that school is out. The image is labelled as human-made with no generative AI, and carries tags for street and urban photography, travel, and photo of the day. The post has drawn modest engagement from photography enthusiasts following the daily suggestion prompt.

A writer let the Opus 5.5 model run for six hours on a single prompt to generate visual interpretations of Italo Calvino's Invisible Cities, the 1972 novel of imagined cities narrated by Marco Polo. The resulting images are being shared and discussed by readers fascinated by how well the model captured Calvino's surreal, dreamlike cityscapes from so little direction.

Nikon Instruments has stripped the first-place winner of its Small World in Motion microscopy competition of the title after determining the winning video was generated with artificial intelligence, violating the contest's rules prohibiting AI-generated entries. The decision is fueling debate in the science and imaging communities about how AI is creeping into scientific visualization and how competitions should verify entrants' work.

As AI-generated images flood the internet, art made by people is reportedly gaining a 'human premium', with buyers and audiences placing extra value on work they know came from a human hand. Commenters are debating whether this signals a lasting market shift or simply nostalgia, and what it means for artists' livelihoods.

As AI-generated images and music flood the market, work made by people is increasingly commanding a price and prestige premium, with buyers paying more for the assurance that a human created it. The shift is prompting artists and collectors to treat human authorship as a mark of value, while galleries and platforms weigh how to certify originality.

SynthID Detector, a tool for identifying content watermarked with Google DeepMind's SynthID technology, is drawing attention. The tool scans text, images, audio and video for invisible SynthID watermarks, letting users check whether material was generated by AI. Commenters are discussing how well detection works and how widely watermarking has been adopted across Google's models and partner tools.

A popular technology review team put AI-based photo editing tools to the test, having professionals try out fake or AI-generated Photoshop features and compare them against the real software. The stunt highlights how quickly AI image editing is evolving and how closely the imitators now rival Adobe's industry standard. Viewers are debating how well the AI tools performed and what this means for the future of professional photo editing.

A judge in Seville, Spain, has issued the first conviction sentence for using generative artificial intelligence to create fake nude images of former schoolmates, with three men receiving prison terms. The ruling, announced by Spain's judiciary, is being flagged as a landmark case for the criminal liability of deepfake abuse targeting women and girls.

A winner of Nikon's Small World microscopic photography competition has been disqualified after it emerged the entry used generative AI. The competition, one of the most prestigious in scientific imaging, requires images to be genuine photographs captured through microscopes. The case has reignited debate about how photo contests verify entries as AI-generated images become harder to detect.

Linum AI's Pyramid Flow research outlines a method for training text-to-image models without relying on a variational autoencoder (VAE), instead using a joint input-output training approach. The technique is being discussed as a potential simplification of generative image pipelines, removing a component that has traditionally added complexity and memory overhead to diffusion model training.

Nikon has stripped the winner of its Small World in Motion competition of the title after determining the winning microscope video used generative AI. Outlets including PetaPixel, Gizmodo and the New York Times are covering the decision, which has sparked debate about AI use in scientific imaging contests and whether competition rules are keeping pace with generative tools.

A black-and-white photograph of the Arctic Cathedral in Tromsø, Norway, is circulating under a photo suggestion challenge inviting images on the theme of 'geometric shapes'. The picture highlights the church's angular architecture, tagged as travel and photography content and explicitly marked as free of AI-generated imagery. Engagement is modest, but the theme-based prompt format is drawing participation from photographers.

A winning entry in Nikon's Small World in Motion microscopy competition has been disqualified after judges found it involved AI use, handing the title to a new winner. The case has drawn attention to how image-making competitions police AI-generated or AI-assisted content, and how contest rules are adapting to generative tools in scientific photography.

Nikon's Small World in Motion photomicrography competition has a new winner after the original first-place entry was disqualified for using AI-generated content. The disqualification has drawn attention to how scientific imaging contests verify submissions and whether AI tools should be allowed in microscopy photography, with Ars Technica covering the change at the top of the results.

ArtCraft, a new AI tool aimed at giving artists more control over generative image creation, is drawing attention online. The product pitches itself as controllable AI for creative work, letting artists direct outputs rather than accept whatever the model produces. Early reactions are focused on whether such tools can genuinely serve working artists rather than replace them.

Photographers in Victoria, British Columbia are sharing images of storm waves and driftwood along Stans Beach on the James Bay stretch of the Dallas Road shoreline. The photos, posted under the SilentSunday photography hashtag, capture rough seas battering the coast in the yyj area, with contributors emphasizing the images are not AI-generated.

A winning entry in Nikon's Small World in Motion microscopy competition has been disqualified after judges determined it involved AI use, according to Ars Technica. The case is drawing attention as science imaging contests grapple with how to handle generative AI tools, with commenters debating whether AI-assisted processing should bar entries from competitions honoring genuine microscope footage.

Nvidia has integrated its RTX video upscaling technology into ComfyUI, the popular open-source interface for AI image and video generation workflows. The move gives users of the community-driven pipeline access to GPU-accelerated upscaling to improve the resolution and quality of AI-generated video. The pairing of a major hardware vendor's proprietary tech with a grassroots open-source tool is drawing attention from the AI generation community.

Nikon disqualified the winner of its Small World in Motion microscopic video competition after finding the entry used generative AI, which the contest rules prohibit. The Verge reported the decision, and it has sparked debate about how photography and imaging competitions verify entries and enforce rules against synthetic imagery as AI tools become more capable and harder to detect.

Black Forest Labs is drawing attention with FLUX 3 Image, the latest entry in its FLUX family of text-to-image models, showcased on the company's model page. Discussion online is focusing on the new release, though details about its capabilities, pricing or benchmarks are thin so far.

A microscope video that won a contest has been disqualified after organizers learned it was created with generative AI tools. The case has sparked debate about how imaging competitions should verify entries and whether AI-generated scientific visuals undermine the integrity of microscopy contests.

Nikon has ruled that a prize-winning photograph, which had drawn a backlash, was in fact generated by artificial intelligence rather than taken with a camera. The finding raises fresh questions about how photography competitions verify entries and police AI use. Commentators are debating whether contest rules and judging processes can keep pace with increasingly realistic synthetic images, and what this means for trust in photographic awards.

Officials at the Big Fresno Fair in California say rumors that Immigration and Customs Enforcement agents were active at the fairground are false. According to the fair's leadership, the claims circulating in the community are based on misinformation and artificially generated images, and no ICE enforcement action has taken place at the event. Fair organizers are urging visitors not to spread unverified claims.

Dramatic images of stormy seas, showing waves crashing under thunderous skies, are being shared across social platforms. The images are generated with open-source AI tools running locally on Linux hardware with AMD graphics cards, and their creators highlight both the moody seascapes and the free software workflow behind them. The combination of storm season visuals and accessible AI art tools is drawing engagement from weather and digital art communities alike.

As AI-generated images flood the art market, human-made works are gaining a 'human premium' — buyers paying more for art created by people. The BBC examines whether this premium will last, or whether AI output will erode the value of human craftsmanship as the technology improves and becomes more widespread.

Researchers at the University of Colorado Boulder have created an improved method for detecting AI-generated deepfake videos and images. As generative AI tools make fake media increasingly realistic and widespread, the technique aims to help researchers, platforms and the public distinguish manipulated content from authentic recordings, addressing growing concerns about misinformation and fraud.