Discover the Best AI Tools

We test the top AI tools for writing, video, images & music — so you don't have to.

Explore AI Tools

OpenAI's Decisions API Is Now Live for Everyone — and It Was Built to Catch a Startup

Image
  Most AI answers arrive as essays. OpenAI's new Decisions API doesn't write a word — it answers yes or no, picks from your categories, or scores on your rubric, in 100 to 180 milliseconds. It went generally available this week, and the backstory is the best part: OpenAI's own product lead, Nikunj Handa, reportedly admitted the product wasn't on the roadmap four weeks ago. A small startup forced the sprint. Yes or no, in 100 milliseconds The API runs on GPT-6 Luna. Developers send text or images plus typed questions and get machine-readable answers in three shapes: a probability that a condition is true, a pick from a fixed list with a probability for each, or a score on rubric levels you define. Input costs $0.10 per million tokens, and output tokens are free. It launched as a public beta on October 6 (after a limited DevDay preview in late September), and the general release adds voice and visual input plus geographically targeted inference. Early customer Speak, the...

Personal AI Just Launched MODEL-4 — Your Phone Carrier Might Soon Sell You AI

Illustration of a phone plan becoming an AI agent — Personal AI MODEL-4

 
Personal AI logo
Personal AI just handed your phone carrier a brand-new product to sell: intelligence itself. On October 20, 2026, the company unveiled MODEL-4, an "operating system" that lets telecom operators deliver personal intelligence to every subscriber line as a network service — billed right alongside talk, text, and data.


What MODEL-4 actually is


MODEL-4 brings together compute, network, and orchestration infrastructure so that a carrier can attach a personal agent to every SIM on its network. Each deployment combines Personal AI's proprietary memory, a unique identity for every subscriber, on-network inference, and multi-modal agents delivered on every line. The launch is backed by an ecosystem of partners: NVIDIA, Microsoft, Rafay, Cisco, and Telegent.


The network decides where AI runs


The pitch is that the network itself becomes the computer. Deutsche Telekom board member Rodrigo Diehl put it this way: the network will understand context and decide where an AI workload is best executed — on the device, at the edge, in a sovereign data center, or in the cloud. To that end, Personal AI already runs small language models — from hundreds of millions to 2 billion parameters — at telco edge sites, citing NVIDIA research that validates them as 20x+ more cost-efficient than large language models for agentic AI.


Why it matters


Carriers have spent two decades watching the value migrate to apps riding on top of their pipes. MODEL-4 is a bid to reverse that: if a personal agent with long-term memory becomes a line item on your phone bill, the carrier owns the billing relationship for AI the way it already does for data. It also fits two trends we have been tracking — AI agents moving out of chat windows and into infrastructure, and the open-source AI models wave that keeps getting cheaper and more capable.


The take


CEO Suman Kanuganti summed up the ambition in one line: "Carriers connected the world through talk, then text, then the internet. Intelligence is the next connectivity, and carriers are the ones best placed to deliver it." If MODEL-4 lands the way Personal AI hopes, your next "unlimited plan" might bundle an agent that remembers everything — and your carrier quietly becomes your AI provider.

Comments

Popular posts from this blog

AI Tools Every Accountant Should Use in 2026

Top 5 Best AI Video Generation Tools in 2026

7 Best AI Website Builders in 2026