Urgent.News

What's breaking now, across thousands of outlets.

AI

Instinct users say they've noticed a slowdown with the buzzy AI agent

The buzzy personal AI assistant startup is competing with Meta's Muse app for users. Meta has access to much more compute.

Users of Instinct, a popular AI personal assistant, are reporting a slowdown in the app's speed. The startup, which is competing with Meta's Muse app for users, faces challenges with the massive compute required for its AI agent. Instinct users have noticed a slowdown in the app's speed when answering questions and completing tasks.

One user, Rex Lowther, said the app sometimes "went home for the night" and failed to ping him when an email arrived. Instinct's founder, Noah Shinn, acknowledged the slowdown, stating that the app's startup status hinders its ability to scale computing resources effectively. While Instinct and Muse have similar features, Meta's Muse has an edge due to its ability to scale computing resources without outages, as noted by a computer science professor at Georgia Tech.

Instinct, a free app that depends on investor funding, is still not fully rolled out to users. The competition between Meta and Instinct has also spilled over onto social media, with Meta's AI chief comparing the two apps' speeds.

Written by urgent.news from Business Insider's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

This story

This is one outlet's version. Read the fullest account.

Read the original at businessinsider.com →

More in AI

OpenAI Tests Sponsored Agents in ChatGPT Ads, Opening Brand Chats From Ads

OpenAI is testing Sponsored Agents , a ChatGPT advertising format that lets users open a dedicated conversation with a brand after clicking an ad.

  • OpenAI testing Sponsored Agents in ChatGPT ads.
  • Allows users to open brand-specific conversations.
  • Currently available to select advertisers in multiple countries.

Running a Jev-Style Decision Model on One TPU v6e: What Fits, What It Costs, and What Changes From a GPU

This article provides a step by step guide to running a Jev-style decision model on one Google Cloud TPU v6e chip with Gemma 4 and vLLM, and compares it with the same read on one NVIDIA L4 GPU.

  • Jev-style decision model runs on Google Cloud TPU v6e using Gemma 4 and vLLM
  • TPU serves 12B models in E2B, E4B, and 12B configurations in bf16 precision
  • GPU can handle same models in bf16 precision, but 26B-A4B and 31B exceed TPU memory

More from Thursday 24 September →