Apple has introduced LensVLM, built on Qwen3.5-9B-Base, which lets vision-language models scan compressed text-as-image inputs and selectively expand only relevant portions using learned tools rather than reading everything at full resolution. The model reportedly matches full-text accuracy at 4.3x compression and beats retrieval and compression baselines up to 10.1x compression across seven QA benchmarks, plus document and code tasks.
huggingface.co
· 2026-09-23
A commentary piece argues that DHH's Omarchy Linux distro is being built and marketed around heavy use of large language models and agentic AI workflows, following a lengthy podcast interview in which DHH discussed his use of AI tools. The piece also notes Alibaba's Phoronix-reported Qwen Book announcement, which frames plans for Omarchy to become the operating system for Alibaba's proposed agentic computer hardware, and mentions Michael Dell's donation and DHH's on-air praise of the Dell XPS.
mikael.pika.page
· 2026-09-23
Google unveiled Custom Feeds at its Made on YouTube event, a feature letting users type natural-language requests to generate a personalized homepage feed. The tool uses large language models to interpret prompts like genre or mood combinations and build a matching set of video suggestions. YouTube tested the feature earlier this year and plans a full rollout to all US users across mobile, desktop and TV.
wired.com
· 2026-09-23
A technical blog post demonstrates how the buzzy AI concept known as 'Jev' can be replicated using an open-source Qwen3-0.6B model and roughly 25 lines of Python code. The example loads the model, prompts it to classify an email as legitimate, spam, or phishing, and extracts probability scores directly from the model's output logits rather than relying on an external API or specialized training pipeline.
nobodywho.ai
· 2026-09-23
An investigation identified more than 80,000 AI relay servers that allow users in China to obscure their location and identity while querying frontier large language models hosted in the US. The relays effectively let these users bypass geographic and access restrictions placed on advanced AI systems.
darkreading.com
· 2026-09-22
TypeSafe's Jev, a rapidly adopted AI classification model, has become the fastest-growing model in Vercel's AI Gateway history. A commentator argues OpenAI could quickly replicate Jev's core approach and bundle it directly into its own models and agents, since OpenAI's LLMs already function as implicit classifiers that could be retrained for this purpose.
arcturus-labs.com
· 2026-09-22
A study examined how assigning different weights to training sequences affects loss reduction in language models of varying sizes, using both proprietary and open-weight models. It found that the relationship between sequence weight and learning is not linear across scale: small models learn general patterns regardless of weighting, medium-scale models learn patterns roughly proportional to their assigned weights, and large models again learn broadly regardless of weighting.
blog.janestreet.com
· 2026-09-21
In a new essay, writer Cory Doctorow argues that people project intention and meaning onto AI systems like Anthropic's Claude much as religious believers see purpose in a sunset. He suggests this instinct to read agency into statistical text generators is a kind of collective delusion rather than evidence of genuine understanding inside the machine.
pluralistic.net
· 2026-09-21
Heretic is a new tool designed to remove safety restrictions and refusal behaviors from language models, allowing them to follow user instructions without the typical content filters. It targets the alignment layers that model developers add to prevent certain outputs, effectively 'uncensoring' these systems for users who run them.
heretic-project.org
· 2026-09-21
An AI researcher argues that the perceived intelligence of chat-based language models isn't a property of the models themselves but an illusion created in users' minds, similar to how psychics use cold reading techniques. The author, who wrote a book called 'The Intelligence Illusion,' contends that LLMs are purely statistical text-prediction systems with no reasoning capability, and that user belief in their intelligence stems from the same psychological mechanisms exploited by cold readers.
softwarecrisis.dev
· 2026-09-20
The United Nations unveiled the UN System Data Commons, a new platform built on Google's open-source Data Commons technology that lets people query UN statistics using plain-language questions and supports the Model Context Protocol so AI systems can pull data directly. It replaces the older UNData portal, which relied on manual browsing rather than conversational search. The announcement came alongside a UNICEF study showing leading chatbots answered development-data questions correctly only about 21% of the time.
techcrunch.com
· 2026-09-17
IEEE Computer Society highlights that most AI products rely on large language models for reasoning and generation, but these models depend heavily on retrieving relevant background information before producing answers. When that retrieved data is outdated, weak, or irrelevant, the resulting output suffers regardless of the model's sophistication.
wpcms.computer.org
· 2026-09-17
GoKawiil