Tech News
← Home  ·  All topics

Ai

5489 GoKawiil briefs on this topic · RSS for ai

Trump's White House 'restoration' photo flagged as AI-edited by Google detector

President Trump posted a before-and-after photo on Truth Social showing a White House column touted as freshly restored, but Google's SynthID Detector flagged the 'after' image as created or edited with OpenAI tools. Reddit users noted the two photos share near-identical angles, window reflections and cardboard placement, but have mismatched and uneven numbers of steps. Other photos Trump posted of the White House facade and heliport did not trigger the AI watermark detection.

Meta and Apple Partner to Enable Open-Weight AI Models on Macs

Meta and Apple have struck a partnership allowing developers to run Meta's open-weight AI models locally on Mac hardware. The collaboration gives developers an alternative to relying on cloud-based computing power for running advanced AI systems.

CoreWeave's 5-year credit default swaps top 1,000 basis points, signaling distress

CoreWeave's credit default swap spread spiked above 1,140 basis points before settling near 1,088, up 33% in a week and 177% over the past year, as bond yields on its 2031 notes climbed to nearly 16%. The neocloud company, which has amassed over $51 billion in debt and obligations for AI infrastructure, saw its stock fall further Thursday after a 15.5% drop the prior day, while peers Nebius and Iren also declined.

New 'decision model' AI systems tested on classic Trolley Problem ethics dilemma

A developer ran the Trolley Problem thought experiment through Jev and twelve other 'decision models,' including OpenAI's Luna and eleven other local models, plus a baseline called Laya. These decision models are a new category of AI that output discrete choices with confidence probabilities rather than text, and are marketed as much faster and cheaper than large language models while retaining comparable reasoning ability. The experiment repeated the classic ethical dilemma in multiple phrasings across all models to compare their responses.

Krea releases Krea 3 image model, claims 4x speed over Nano Banana Pro

Krea has launched Krea 3, a new AI image generation model that the company says runs four times faster than Google's Nano Banana Pro. The release is accompanied by a set of example multi-image composition prompts showing the model combining and editing elements from several reference images into a single output.

Silicon Valley AI super PACs spend heavily but avoid mentioning AI in ads

Fast Company reports that tech-industry-backed super PACs are spending tens of millions of dollars to influence midterm elections, yet their political advertisements largely avoid discussing artificial intelligence. The PACs are funded by donors with a stake in favorable AI policy, but their messaging focuses elsewhere.

Apple's HomeView smart display arrives Nov. 13 with a 7-inch screen

Apple unveiled HomeView, its first smart home display with Siri integration, on Tuesday, with availability starting Nov. 13. CNET's reviewer, who has tested Echo Show and Nest Hub devices for years, got hands-on time with HomeView and found its design similar to Amazon's Echo Show 11, but with a notably smaller 7-inch screen and a stationary base rather than a rotating one.

Bluesky expands Attie AI tool to let users build and share mini social apps

Bluesky's Attie AI tool, which debuted earlier this year as a way to build custom feeds, has been expanded to let users create small social apps such as games, polls, and dashboards. These apps can then be shared as standalone websites. CEO Toni Schneider described the tool at TechCrunch Disrupt as a form of 'social vibe coding' for communities.

TSMC reports record quarterly profit amid surging AI chip demand

TSMC posted its sixth consecutive quarter of record profit and forecast fourth-quarter revenue up 58% year-over-year, driven by demand for advanced chipmaking. Executives say wafer capacity is stretched thin as clients including Nvidia, AMD, Apple, Google, Amazon, Microsoft, Meta and OpenAI compete for production slots.

Google adds Gemini-powered scam detection to Android Messages and calls

Google is rolling out AI-based scam defenses on Android, including cloud-based Gemini analysis of suspicious conversations in Google Messages and on-device Gemini Nano checks on unfamiliar links. The company says its existing detection already flags over 5 million suspected scam calls and texts monthly, and the new cloud protections are opt-in, while on-device chat notification detection expands beyond Pixel to more Android partners later this year.

Apple, Amazon and Microsoft roll out AI-focused devices ahead of holiday season

Apple unveiled HomeView, a $299 smart home hub with facial recognition and an AI-enhanced Siri. Amazon replaced its Fire tablet line with three new Alexa-branded tablets starting at $500 that emphasize voice control and on-screen intelligence features, while Microsoft introduced an AI-chip-powered Windows laptop priced from $2,599.

Robotics industry grapples with scarce training data as AI scaling model falters

AI language models have advanced largely by training on ever-larger text datasets, with dataset sizes reportedly doubling roughly every six months since the early 2010s, according to Epoch AI. But researchers note there is no equivalent large public corpus of robot training data, unlike the estimated 300 trillion tokens of usable text available for language models. This gap has made sourcing and defining robot training data a central open problem for the field.