D.O.T.S AINEWS
Live
OpenAI announces GPT-5 with unprecedented reasoning capabilitiesGoogle DeepMind achieves breakthrough in protein folding for rare diseasesEU passes landmark AI Safety Act with global implicationsAnthropic raises $7B as enterprise demand for Claude surgesMeta open-sources Llama 4 with 1T parameter modelNVIDIA unveils next-gen Blackwell Ultra chips for AI data centersApple integrates on-device AI across entire product lineupSam Altman testifies before Congress on AI regulation frameworkMistral AI reaches $10B valuation after Series C funding roundStability AI launches video generation model rivaling SoraOpenAI announces GPT-5 with unprecedented reasoning capabilitiesGoogle DeepMind achieves breakthrough in protein folding for rare diseasesEU passes landmark AI Safety Act with global implicationsAnthropic raises $7B as enterprise demand for Claude surgesMeta open-sources Llama 4 with 1T parameter modelNVIDIA unveils next-gen Blackwell Ultra chips for AI data centersApple integrates on-device AI across entire product lineupSam Altman testifies before Congress on AI regulation frameworkMistral AI reaches $10B valuation after Series C funding roundStability AI launches video generation model rivaling Sora

Breaking

2 articles tagged "Breaking"

Perplexity Launches 'Personal Computer' — An Always-On AI Agent for $200 a Month
Breaking2 min read

Perplexity Launches 'Personal Computer' — An Always-On AI Agent for $200 a Month

Perplexity AI has announced a product it is calling the Personal Computer: a subscription service at $200 per month that provides users with a persistent, always-on AI agent capable of browsing the web, executing long-horizon research tasks, managing files, writing and running code, and operating third-party apps on behalf of the user — without the session resets that limit current chatbot interfaces. The product is positioned as a direct competitor to OpenAI's Operator and Anthropic's upcoming agentic offerings, but with a key differentiator: Perplexity's long-standing strength in real-time web search is baked natively into the agent, making it particularly effective for research and information-dense tasks. Early access users in the beta program describe a qualitative jump in what an AI system can accomplish autonomously over multi-hour spans — including completing competitive analyses, preparing briefing documents, and monitoring news on specific topics without requiring continuous user interaction. The $200/month price point is aggressive for a consumer product but positions it below enterprise agent platforms, which typically run $500–2,000/month for comparable autonomous task capability.

James LiuJames Liu2 min read
Meta Delays 'Avocado' AI Model After Internal Benchmarks Show It Trails Google and OpenAI
Breaking

Meta Delays 'Avocado' AI Model After Internal Benchmarks Show It Trails Google and OpenAI

Meta has postponed the release of its next flagship AI model, internally code-named Avocado, after evaluation results showed the model performing below Google's Gemini 2.5 Pro and OpenAI's GPT-4.5 on key capability benchmarks — a result the company considers unacceptable given its public positioning of Llama as a frontier-competitive open-weights alternative. Sources familiar with Meta's AI org say the delay reflects both disappointing benchmark performance and internal disagreement about the model's readiness for open-weight release, with some researchers arguing that releasing a model that demonstrably trails closed competitors would damage the Llama brand's authority in the open-source community. The delay is notable coming just weeks after Meta CEO Mark Zuckerberg told investors the company planned to release Avocado in Q1 2026. It also coincides with a broader reset in expectations about the pace of open-source model progress: DeepSeek's R2, widely anticipated as a potential leap past Western models, has also reportedly been pushed back after internal testing surfaced reasoning reliability issues.

Sofia Reyes