Archive: August 2015

In recent years, much of the research interest in large language models (LLMs) such as OpenAI’s autoregressive GPT has shifted from what these models can do to how they do it. While LLMs have demonstrated impressive prediction consistency with factual knowledge, their computations remain opaque. Knowing how and where such factual associations are stored and retrieved and improving understanding of the mechanisms underlying autoregressive knowledge represen

Low-Rank Adaptation (LoRA) is a novel technique introduced by Microsoft in 2021 for fine-tuning large language models (LLMs). LoRA is an efficient adaptation strategy that introduces no additional inference latency and substantially reduces the number of trainable parameters for downstream tasks while maintaining model quality. Although LoRA was initially proposed for LLMs, it also can be applied elsewhere. Inspired by the Stable Diffusion paper published

Create a watch on this topic Receive the next developments if any, with an AI watch configured from this article. Create my free watch Prometheus, the AI physical startup co-founded late 2025 by Jeff Bezos and Vik Bajaj, announced on June 11, 2026, a second round of $12 billion at a valuation of $41 billion, according to TechCrunch. The syndicate includes Bezos personally, JPMorgan Chase, Goldman Sachs, BlackRock, as well as DST Global and Arch Venture Par

Credit: Armand Valendez from Pexels Apple is trying to bring machine learning to Motor City. The smartphone giant opened its Apple Manufacturing Academy in Detroit on Tuesday to offer free workshops on artificial intelligence and advanced manufacturing to small and medium-sized businesses. Apple stands out as one of the companies most dependent on manufacturing as well as consumers in China. The new academy is a part of Apple's broader $500-billion U.S. in

Illustration via Getty / Futurism If you’re puzzled as to why OpenAI’s ChatGPT seems a little stupider lately, you’re not alone. “It straight up told me I was dying out of nowhere when I asked about a hot spot on my arm,” one Reddit user complained. And that’s without getting into how the bot has been going haywire if you ask if for a seahorse emoji, or a list of NFL teams that don’t end with the letter “s,” or getting lawyers fired by making up fake case

Getty / Futurism Our society has only recently come to terms with the fact that maybe placating young children by getting them hooked on touchscreen devices with unfettered access to the internet was bad for their brains. Now, with the rise of human-like AI chatbots, a generation of “iPad babies” could seem almost quaint: some parents are now encouraging their kids to talk with AI models, sometimes for hours on end, The Guardian reports. Others are using t

Anthropic finds 94 cases where open-source projects drop security controls without warning Software can expose complex functionality to unauthenticated users even when that functionality is disabled or unnecessary. Anthropic’s Frontier Red Team used Claude to examine 25 open-source projects and reported 493 issues involving unnecessary exposure, conflicting settings, and code running before security checks. Attack surface reduction means limiting the parts

Illustration by Tag Hartman-Simkins / Futurism. Source: Getty Images If you’ve spent any time around kids lately, you’ve probably heard about “brain rot.” Named Oxford Word of the Year in 2024, it’s defined as the “supposed deterioration of a person’s mental or intellectual state, especially viewed as the result of overconsumption of material (now particularly online content) considered to be trivial or unchallenging.” As it turns out, it’s not just human

Ransomware gang talked AI agents into doing much of its dirty work The group behind the Aur0ra ransomware has been using AI agents to move through victim networks, collect credentials and work its way up to domain administrator access, according to Sophos. AI-assisted intrusion workflow (Source: Sophos) The company, which tracks the group as GOLD OGRE, recovered artifacts from the Cursor AI coding assistant in late August that show the attackers handing th

Getty / Futurism Children’s toymaker FoloToy says it’s pulling its AI-powered teddy bear “Kumma” after a safety group found that the cuddly companion was giving wildly inappropriate and even dangerous responses, including tips on how to find and light matches, and detailed explanations about sexual kinks. “FoloToy has decided to temporarily suspend sales of the affected product and begin a comprehensive internal safety audit,” marketing director Hugo Wu to

The Raine Family Content warning: this story includes discussion of self-harm and suicide. If you are in crisis, please call, text or chat with the Suicide and Crisis Lifeline at 988, or contact the Crisis Text Line by texting TALK to 741741. OpenAI has shot back at a family that’s suing the company over the suicide of their teenage son, arguing that the 16-year-old used ChatGPT incorrectly and that his tragic death was his own fault. The family filed the

For towns and cities across the country looking for ways to tell data center developers that their projects aren’t welcome in their communities, the fine folks of Salem, Oregon seem to have found a very efficient way to get their message across: Pull up to public hearings with a guillotine. Last month, representatives from Verrus, a California-based data center development company, made the trek up to the city of 180,000 to pitch a planned $5.1 billion ser

Free-energy trends in Z-SINDy inference. Credit: Physical Review Research (2025). DOI: 10.1103/4d98-tdlp A study by University of Hawaiʻi researchers is advancing how we learn the laws that govern complex systems—from predator-prey relationships to traffic patterns in cities to how populations grow and shift—using artificial intelligence (AI) and physics. The research, published in Physical Review Research, introduces a new method based on statistical mech

Getty / Futurism Since the start of 2025, at least 30 cities have canceled their contracts with Flock Safety, the AI surveillance company whose CEO wants to end all crime within the decade by blanketing the country in ever-watchful security cameras. That startling figure comes courtesy of NPR, which reports that concerned activists are putting mounting pressure on cities to cut ties with the company. “We are seeing a lot more momentum,” Will Freeman, a Col

Meta / Futurism Meta may have sold seven million of its Ray-Ban smart glasses in 2025 alone — but likely didn’t anticipate the outpouring of criticism when a recent investigation by Swedish newspapers Svenska Dagbladet and Göteborgs-Posten revealed that Meta’s subcontracted data annotators in Nairobi, Kenya, could’ve been watching users through their glasses’ cameras as they went to the bathroom or had sex. The damning revelations shed light on the AI indu