ChatGPT now has its own web search engine
OpenAI is challenging Google in its own search turf
Large language model articles
The promise (and pitfalls) of NPCs powered by AI in video games
The mods are out there, but are they really more than just a novel attraction?
Meta's AI chief Yann LeCun calls AI apocalypse fears "complete B.S."
The Godfather of AI says current LLMs aren't as smart as house cats
Foxconn to open Mexico plant for mass production of Nvidia Blackwell superchips
Foxconn and Nvidia are also working together on a new AI supercomputer in Taiwan
New algorithm promises to slash AI power consumption by 95 percent
How hard will Nvidia fight adoption since it requires a competitive hardware redesign?
OpenAI may hike ChatGPT subscriptions to $2,000 for next-gen AIs, but terms remain unclear
New AI model "Strawberry" could launch as early as this fall with more "human-like" analytical abilities
IBM debuts new AI hardware with Telum II CPU and Spyre Accelerator chip
Big Blue is all about intelligent workloads and generative AI acceleration these days
Reddit stands firm against AI companies scraping content for training without paying
"It's free as in speech, not free as in beer"
Meta AI Studio lets you clone your personality into a chatbot
When life seems unreal, insert more articial intellegence products
OpenAI is replacing GPT-3.5 with new GPT-4o mini model for free users
It's over 60% "cheaper" to process than its predecessor
Microsoft is developing an AI model for Excel spreadsheets, enabling natural language prompts
Productivity booster or job killer?
Survey shows many people believe AI chatbots like ChatGPT are conscious
I don't think, therefore I am not?
Transhumanist author predicts artificial super-intelligence, immortality, and the Singularity by 2045
Dystopian Kurzweil: As Big Tech continues frantically pushing AI development and funding, many users have become concerned about the outcome and dangers of the latest AI advancements. However, one man is more than sold on AI's ability to bring humanity to its next evolutionary level.
Cloudflare offers one-click solution to block AI bots
Why it matters: There is a growing consensus that generative AI has the potential to make the open web much worse than it was before. Currently all big tech corporations and AI startups rely on scraping all the original content they can off the web to train their AI models. The problem is that an overwhelming majority of websites isn't cool with that, nor have they given permission for such. But hey, just ask Microsoft AI CEO, who believes content on the open web is "freeware."
Energy-efficient AI model could be a game changer, 50 times better efficiency with no performance hit
Cutting corners: Researchers from the University of California, Santa Cruz, have devised a way to run a billion-parameter-scale large language model using just 13 watts of power – about as much as a modern LED light bulb. For comparison, a data center-grade GPU used for LLM tasks requires around 700 watts.
Google defends failing AI search, while promising improvements
AI Overviews struggle with satire, lack of information, and confusing queries
Google search AI Overviews uses terrible information sourcing to give you terrible answers
Google downplays the bad answers, saying they "aren't representative of most people's experiences"
Google is now the world's third-largest data center processor designer
Custom TPUs have driven past growth, and will accelerate the future as well
Complex adversarial attacks can force generative AI services to bypass security filters and limitations
The right text strings can push chatbots to break the law or teach people how to create bombs
Nvidia thinks its GPUs are better for onboard AI than NPUs
Company claims RTX GPUs perform AI tasks multiple times faster than NPUs
Microsoft's Phi-3 Mini boasts ChatGPT-level performance in an ultralight 3.8B parameter package
Why it matters: Advanced AI capabilities generally require massive cloud-hosted models with billions or even trillions of parameters. But Microsoft is challenging that with the Phi-3 Mini, a pint-sized AI powerhouse that can run on your phone or laptop while delivering performance rivaling some of the biggest language models out there.
Apple said to be working on custom AI server chip based on TSMC's 3nm process
Cupertino isn't interested in efficiently running chatbots and LLMs on local devices alone
GPT-4 can exploit zero-day security vulnerabilities all by itself, a new study finds
A hot potato: GPT-4 stands as the newest multimodal large language model (LLM) crafted by OpenAI. This foundational model, currently accessible to customers as part of the paid ChatGPT Plus line, exhibits notable prowess in identifying security vulnerabilities without requiring external human assistance.
Meta's Llama 2 LLM is still prone to hallucinations and other severe security vulnerabilities
It failed most tests put to it by researchers
Los Alamos' Nvidia-powered Venado supercomputer trades accuracy for efficiency with 10 exaFLOPS for AI workloads
The first US super-system designed around Nvidia's Grace Hopper chip architecture