Tech News
← Home  ·  All topics

Ai

5489 GoKawiil briefs on this topic · RSS for ai

AI agents push cybersecurity toward continuous, automated defense cycles

Companies are increasingly deploying automated AI systems that continuously test, patch, and defend software rather than relying on periodic security reviews. The shift comes as AI tools have demonstrated growing skill at discovering software vulnerabilities and circumventing existing cybersecurity defenses.

DeepSeek open-sources DeepInfer, its production inference engine

DeepSeek has released DeepInfer, the inference framework it uses internally to serve its large-scale production systems, as an open-source project. The initial release supports the DeepSeek-V4.1-Flash model and includes optimized kernels, parallelism strategies, speculative decoding, asynchronous scheduling, and batch-invariant deterministic inference.

Lightroom mobile adds one-tap generative AI tool to open closed eyes in photos

Adobe has added an Open Eyes Quick Action to the mobile version of Lightroom for iOS, iPadOS and Android. When the app detects closed eyes in a photo, it surfaces an icon that uses generative AI to replace them, drawing on other images of the same person in the user's library to match eye color, size and shape.

DHS orders federal agencies to purge DeepSeek AI products and derivatives

The Department of Homeland Security has directed US federal civilian agencies to remove DeepSeek's AI products from government systems and barred future purchases, including third-party tools built on DeepSeek's models. The order, dated late July but disclosed publicly on October 19, cites the Federal Acquisition Supply Chain Security Act of 2018 and concerns about DeepSeek's China-based servers and potential vulnerabilities. China's Foreign Ministry called the move a politicization of trade and technology.

Report finds 2,100+ AI-generated recipe sites using fake chef personas, including an Emily Blunt deepfake

Ad-supply researcher Decrypt Ads published a report identifying more than 2,100 websites publishing AI-generated recipes under invented chef identities, some promoted using an AI-generated likeness of actress Emily Blunt. Researcher Zach Edwards said he started with about 400 suspicious domains and used technical fingerprinting to trace hundreds more linked to the same operators, who monetize traffic through adtech partners based in Russia and China. The sites push content aggressively on Pinterest, Facebook, and Twitter/X to drive organic traffic to Google search and social platforms.

"The Secret Life of Circuits" book now available on Amazon and major retailers

The author's book, previously sold only through the publisher, is now stocked on Amazon in the US, Canada and multiple European sites, as well as Barnes & Noble and other major retailers. The author notes the book was checked to ensure it contains no AI-generated content.

DHS orders DeepSeek removed from US civilian federal agency systems

The Department of Homeland Security publicized a directive requiring US federal civilian agencies to strip DeepSeek-related technology from their IT systems and halt future procurement of its products and services, including third-party tools built on the Chinese company's technology. Issued July 24, the order covers agencies like Treasury, State and Justice, but exempts the Pentagon, which already restricts the AI models.

Elice Shares Drop Up to 36% in Seoul Trading Debut

South Korean data-center startup Elice saw its stock fall as much as 36% in afternoon trading on its first day as a publicly listed company in Seoul. The sharp decline came despite the company's AI-infrastructure focus, an area that has generally drawn strong investor interest.

Executive describes building an AI version of Jamie Dimon for advice

Financial services executive Ryan Sparks said he created an AI persona modeled on JPMorgan Chase CEO Jamie Dimon to simulate the kind of challenging conversations he once had with Dimon and his operating committee earlier in his career. Sparks described the tool as part of a broader idea of using AI to build a virtual 'board of directors' that executives can consult to pressure-test decisions, rather than just using AI for routine productivity tasks.

Nvidia backs four German AI startups as part of $10B European investment push

Nvidia announced investments in four German startups—Agile Robots, Avelios Medical, Mirelo and Reflex Aerospace—at the Invest in Germany Summit in Berlin, part of a nearly $10 billion commitment to companies, universities and research institutions across Europe. The company did not disclose the size of the individual investments. The announcement coincided with remarks from Chancellor Friedrich Merz urging German businesses, especially small and medium-sized firms, to adopt AI and make greater use of their existing data.

NYC Mayor Mamdani criticizes federal inaction on AI, calls for local pushback

At a public event in Manhattan hosted by the AI Now Institute, Mayor Zohran Mamdani said the federal government has failed to regulate artificial intelligence, leaving cities to respond on their own. He pointed to his administration's recent actions, including a ban on AI use in New York City elementary and middle schools and a policy requiring real estate listings to disclose AI-generated photos or videos.

Baseus launches U2 Pro security camera with motion-tracking spotlight

Baseus has released the Security PrimeSentinel U2 Pro Outdoor 3K Dual Camera, which uses a motorized spotlight to follow moving people or animals across a property. The device combines a fixed 3K camera with a second 3K camera that pans, tilts and zooms 360 degrees, plus full-color night vision and AI-based motion alerts. It's priced at $269.99 in the US, available now, with European sales starting October 15th at €269.99.