Homechevron_rightArchiveschevron_rightTechnology

Archives

Explore Stories by Date

Today

OpenAI launches three voice AI models that translate 70+ languages and reason through live conversations

OpenAI has introduced three audio models through its Realtime API that enable developers to build voice-powered applications with advanced conversational capabilities. GPT-Realtime-2 brings GPT-5-class reasoning to live interactions, GPT-Realtime-Translate handles real-time language translation across 70+ input languages, and GPT-Realtime-Whisper delivers streaming transcription. Companies like Zillow, Priceline, and Deutsche Telekom are already testing these models to build voice assistants that can reason through requests and take action during conversations.

4 Sources

OpenAI releases GPT-5.5-Cyber to vetted cybersecurity teams, a month after Anthropic's Mythos

OpenAI announced GPT-5.5-Cyber, a specialized AI model for cybersecurity teams to identify vulnerabilities and analyze malware. The limited preview follows Anthropic's Mythos release, with both models showing advanced capabilities in finding software bugs. OpenAI's approach offers broader access through its Trusted Access for Cyber program compared to Anthropic's restrictive rollout.

2 Sources

Yesterday

Apple's AirPods with cameras for AI-powered Siri reach advanced testing after four years

Apple's long-rumored AirPods with cameras have entered an advanced testing stage after four years of development. The camera-equipped AirPods feature low-resolution cameras designed to feed visual data to an upgraded Siri AI assistant, not for photo or video capture. Launch timing depends on Apple's delayed AI-powered Siri, now expected in September.

12 Sources

OpenAI Launches ChatGPT Trusted Contact Feature to Alert Friends During Self-Harm Concerns

OpenAI introduced a new ChatGPT safety feature called Trusted Contact that alerts a designated person when users express serious self-harm concerns. The optional feature involves human review and was developed with mental health experts amid growing legal pressure over suicide-related incidents involving the chatbot.

6 Sources

Anthropic Mythos Discovers 271 Firefox Vulnerabilities, Including Bugs Hidden for 15 Years

Mozilla deployed Anthropic Mythos to uncover 271 security flaws in Firefox, resulting in 423 total bug fixes in April—more than 13 times the monthly average. The AI model found decade-old vulnerabilities that traditional fuzzing missed, though skeptics question whether the breakthrough lies in the model itself or Mozilla's custom harness system.

4 Sources

Amazon Bedrock enables AI agents to pay with stablecoins through Coinbase and Stripe integration

AWS launched Amazon Bedrock AgentCore Payments, partnering with Coinbase and Stripe to let AI agents autonomously purchase services using stablecoins. The system uses Coinbase's x402 protocol to process micropayments for APIs, data feeds, and web content, with plans to expand to flight bookings and hotel reservations.

4 Sources

Dragon Quest creator envisions AI characters as empathetic friends, not just helpful tools

Yuji Horii, creator of Dragon Quest, believes AI can make JRPG characters feel more human and provide empathetic support to players. Square Enix is partnering with Google to introduce Chatty Slimey, an AI chatbot companion in Dragon Quest X that players can talk to about game-related topics and personal concerns. Horii hopes to expand AI companions to other series too.

2 Sources

xAI Becomes SpaceXAI as Elon Musk Merges AI Company with SpaceX in Bold Infrastructure Play

Elon Musk's AI company xAI is being dissolved as a standalone entity and merged with SpaceX to form SpaceXAI. The move follows a deal leasing the Colossus 1 supercomputer to Anthropic and signals Musk's push toward orbital data centers. Trademark filings reveal plans for satellite-based AI infrastructure spanning over 220,000 Nvidia GPUs.

2 Sources

Roche acquires PathAI for up to $1.05 billion to strengthen AI diagnostics and digital pathology

Swiss pharmaceutical giant Roche announced it will acquire U.S.-based AI company PathAI in a deal worth up to $1.05 billion. The acquisition expands Roche's digital pathology capabilities and AI diagnostics tools, building on a partnership that began in 2021. PathAI will join Roche's diagnostics division to accelerate precision medicine and biomarker discovery.

2 Sources

Spotify lets AI agents like OpenClaw and Claude Code create Personal Podcasts from your notes

Spotify launched a new beta tool called Save to Spotify that enables AI agents to generate and upload personalized podcasts directly to user libraries. The command-line interface works with OpenClaw, Claude Code, and OpenAI Codex to turn daily briefings, class notes, and research into private audio content accessible across devices.

5 Sources

Google launches AI-powered Health Coach at $9.99 monthly alongside screenless Fitbit Air tracker

Google is overhauling its fitness ecosystem with three major moves: rebranding the Fitbit app as Google Health, launching its Gemini-powered Health Coach subscription service at $9.99 per month, and releasing the $99 Fitbit Air—a screenless fitness tracker designed to compete with Whoop. The coordinated launch signals Google's strategic shift toward AI-driven wellness guidance.

20 Sources

Meta Launches Business AI on WhatsApp Business to Help Small Businesses in India Automate Support

Meta has introduced Business AI within the WhatsApp Business app for small businesses in India. The AI-powered customer support feature can automate customer interactions, respond to queries around the clock, and drive sales growth. Early pilots show companies reporting 30-40% sales growth within weeks of deployment.

3 Sources

IBM and Yotta partner to deliver agentic AI platform for Indian enterprises and government

IBM and Yotta Data Services announced plans for a strategic partnership to build an agentic AI platform using IBM watsonx Orchestrate deployed on Yotta's Shakti Cloud. The collaboration aims to help Indian enterprises and government organizations scale AI adoption while meeting data residency, security, and regulatory compliance requirements.

2 Sources

AMD launches Instinct MI350P PCIe AI accelerator with 144GB HBM3E to challenge Nvidia's dominance

AMD unveiled the Instinct MI350P, a PCIe AI accelerator card with 144GB of HBM3E memory designed for air-cooled enterprise servers. The dual-slot card delivers up to 4,600 TFLOPS performance and outpaces Nvidia's H200 NVL by roughly 40% in FP16 and FP8 theoretical compute. With support for up to eight cards per system, AMD targets enterprises seeking scalable AI infrastructure without major platform overhauls.

7 Sources

Spotify AI DJ adds French, German, Italian, and Portuguese as it expands to 75+ countries

Spotify is expanding its AI DJ feature to new markets after years of user demand. Premium users in Austria, Brazil, France, Germany, Italy, Portugal, South Korea, and Switzerland can now access the AI-powered personal DJ with new language support for French, German, Italian, and Brazilian Portuguese, bringing the total to over 75 countries.

5 Sources

Get all updates into your inbox 💌, Stay Up-to-date for Free 🥳

Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Spend less time searching for the latest in AI and get straight to action.

The Outpost

News

Tech Powerhouse

AI Tools

About Us Privacy Terms Content