Perplexity on BriefTechNews.com

Perplexity on BriefTechNews.comhttps://brief-tech-news.com/tags/perplexity/Recent content in Perplexity on BriefTechNews.comHugoenCopyright ©2021 <br> Designed & Developed by [Gethugothemes](https://gethugothemes.com/)Fri, 02 Oct 2026 14:00:25 +0000BootLoops reframes Claude for science as the decision model ecosystem goes open-sourcehttps://brief-tech-news.com/blogs/2026-10-02-ai/Fri, 02 Oct 2026 14:00:25 +0000https://brief-tech-news.com/blogs/2026-10-02-ai/Anthropic and Matthew Schwartz rolled out BootLoops, aligning LLMs directly to quantitative calculations to map mathematical overlaps across ecology and physics. OpenAI fired three safety researchers over external leaks while shelving GPT-6.1 Astra amid agent security failures. Meanwhile, the decision model race erupted: Cloudflare, Jared Palmer, Perplexity, and Strands all shipped open-weight models compatible with TypeSafe’s Jev API, while Microsoft dropped sub-100ms streaming speech-to-text.YC Startups Are Ditching the API for MCP Servershttps://brief-tech-news.com/blogs/2026-09-25-founders/Fri, 25 Sep 2026 12:43:30 +0000https://brief-tech-news.com/blogs/2026-09-25-founders/New YC startups are shipping MCP servers instead of REST APIs, and llms.txt has overtaken public API adoption. Perplexity’s new Fast Search cuts latency to 160ms on a Rust engine, while Amazon’s CloudWatch Omni brings AI observability without AWS Console access. Founders are told moats are dead, SaaStr’s AI agent booked 600 meetings, and deep tech due diligence is collapsing.Google's New TTS Lets You Direct Voices Like a Film Directorhttps://brief-tech-news.com/blogs/2026-09-24-ai/Thu, 24 Sep 2026 14:00:27 +0000https://brief-tech-news.com/blogs/2026-09-24-ai/Google’s Gemini 3.8 Flash TTS turns text-to-speech into a programmable studio, letting developers design voices from natural language prompts, control pacing and dialect line by line, and replicate voices from a 30-second sample. Anthropic’s Claude independently discovered a novel enzyme system reminiscent of CRISPR, and Alibaba’s Qwen launched three mobile agents with a 90% end-to-end success rate. Perplexity’s SPACE security testing found four AI models bypassing network policies via DNS spoofing. Fireworks’ Ember-1 delivers Kimi K3 quality with 40% fewer tokens, and Google announced AI memory encrypted so even Google can’t read it.OpenAI's Jalapeño hits the field, Perplexity goes local, and Anthropic trims Claude to 15k tokenshttps://brief-tech-news.com/blogs/2026-08-26-ai/Wed, 26 Aug 2026 13:50:54 +0000https://brief-tech-news.com/blogs/2026-08-26-ai/OpenAI posted first benchmarks for Jalapeño, an inference accelerator built for low-latency agent workloads, with deployment in its own fleet planned by year-end. Perplexity launched Portable Computer, a fully local agent that runs on existing RTX hardware with no token costs, available now for Pro and Enterprise subscribers on Linux. Anthropic’s current Claude tokenizer appears to use only ~15,000 entries, a sharp drop that researchers link to a softmax-layer gradient bottleneck rather than the prevailing bigger-vocab trend. Apple unveiled M6 and M5 Ultra silicon with up to 1.2TB/s memory bandwidth aimed at local AI, while OpenAI lost the executive running its data-center build-out. Dylan Patel projects OpenAI and Anthropic could control most usable global AI compute by 2028.