A South Korean research team just unveiled a set of seven wireless AI rings that translate sign language into text on the fly. Worn on dominant fingers, the stretchy rings track motion and use predictive autocomplete to keep up with conversation speeds, achieving over 88% accuracy. It cuts the bulky cables of older sensor gloves, though missing facial cues remains a challenge.
Google DeepMind just reimagined the mouse pointer with Gemini, letting users point at on-screen content and give natural voice commands like "Fix this" without detailed prompting. The system reads the context around the cursor to turn pixels into actionable entities, and it's already rolling into Chrome and the upcoming Googlebook laptop.
Google just unveiled Gemini Intelligence on Android, turning phones into proactive agents that handle multi-step tasks across apps, clean up messy spoken notes with Rambler, and build custom widgets from text prompts. It rolls out this summer on the Galaxy S26 and Pixel 10 before expanding to watches, cars, and laptops later this year.
Google just unveiled Googlebook, a brand-new laptop category that fuses Android and ChromeOS around Gemini intelligence. It turns the mouse cursor into an active assistant called Magic Pointer for quick contextual actions, lets you generate custom desktop widgets with simple prompts, and seamlessly runs phone apps. Partner devices from HP, Dell, and Lenovo will arrive this fall.
AI developers running persistent agents on MacBooks have found a low-tech fix to keep laptops awake with the lid closed: dummy display plugs. Suggested by developer Will DePue on X, cheap adapters trick macOS into clamshell mode. While free software like caffeinate exists, developers prefer hardware dongles because unplugging is simply the fastest way to let the machine sleep.
Cate Blanchett's nonprofit RSL Media just launched the Human Consent Standard, backed by stars like Tom Hanks and Meryl Streep. The framework lets creators explicitly dictate whether AI systems can use their likeness, voice, or creative work. A verification registry will roll out in June, linking identity permissions directly to web crawlers via robots.txt.
Rivian just rolled out Rivian Assistant, a native voice agent deeply wired into Gen 1 and Gen 2 vehicle hardware. Unlike typical phone-mirroring assistants, it directly controls systems like adjusting ride height or opening the front trunk, while also managing Google Calendar events and drafting texts. Why it matters: It shifts in-car voice from clunky dictation to an integrated operational co-pilot, though it currently requires a Connect+ subscription and only supports English.
Ryan Shea just launched AI IQ, a platform that ditches dense benchmark tables to map frontier AI models directly onto the human IQ bell curve. It tracks intelligence growth over time, compares models across both IQ and EQ metrics, and charts practical intelligence costs. Why it matters: It translates obscure benchmark scores into intuitive human metrics anyone can read at a glance.
Krea just launched Krea 2, its first proprietary foundation model built completely from scratch. Designed specifically for aesthetic diversity and stylistic control, it marks the company's shift from building wrappers on third-party tech to owning the underlying models. Early access is open now.
AI researchers just introduced L2P, a new diffusion model that removes the VAE and latent space to generate images directly in pixel space up to 4K. By skipping latent compression, it preserves fine textures and sharp text without typical quality degradation. The team has open-sourced the training code and a 20GB model checkpoint for 1K resolution.
SenseNova just open-sourced SenseNova-U1.5-8B-MoT under the Apache 2.0 license. Skipping traditional latent spaces and VAE decoders, it crafts native 4K images end-to-end directly in pixel space and handles natural-language image edits. The catch: its 50GB weight footprint means you will need a beefy GPU to run it locally.
Google is reportedly in talks with SpaceX to launch AI data centers into orbit, as Elon Musk pitches space as the cheapest home for compute ahead of a mega-IPO. While orbital servers dodge local backlash and land constraints on Earth, sky-high launch and satellite build costs mean ground facilities still win on price by a mile.
Anthropic just rolled out 20+ MCP connectors and 12 practice-area plugins for Claude, directly targeting legal teams. The tools hook into platforms like DocuSign, iManage, and Harvey, while syncing context across Word, Outlook, Excel, and PowerPoint. Why it matters: legal pros are already Claude's top Cowork users, and this turns the AI into a bespoke junior associate across the entire firm.
Anthropic just doubled rate limits for Claude Code and shipped four major upgrades to Claude Managed Agents at its developer conference. The platform now supports parallel multi-agent workflows, automated memory curation called Dreaming, and a built-in rubric grader that boosted success on hard tasks by 10 points.
Unitree just unveiled the GD01, a transformable manned mecha starting at $650,000. Billed as the world's first production-ready manned mecha, the civilian vehicle weighs roughly 500 kilograms with an operator inside and is capable of transforming. Unitree urged future pilots to operate the machine in a "friendly and safe manner."
OpenAI just launched Daybreak, a cybersecurity initiative pairing frontier models with its Codex agentic platform to discover, validate, and patch software flaws. The company is putting up $1 billion in credits over six months for defenders and open-source maintainers, using tiered access to keep advanced capabilities under human oversight.
Vivago AI just released HiDream-O1 Image, an open-source visual model that bypasses traditional VAEs to generate pixels end-to-end. It pushes 2K resolution while nailing text rendering and complex infographic layouts where other open models stumble. The weights clock in at a hefty 32GB across base and dev versions, meaning you will need beefy GPU hardware to self-host it.
Pixal3D just released its pixel-aligned image-to-3D pipeline, generating 3D assets with remarkable geometric precision and texture fidelity. By explicitly projecting 2D pixels into 3D structure rather than relying on loose attention cues, it solves long-standing alignment issues. Why it matters: 3D creators now get assets that actually match the input image close to true reconstruction quality.
Google just launched REPLIQA, a joint initiative by Google Quantum AI and Google.org to apply quantum tech and AI to life sciences. The effort commits $10M to five universities, including Harvard and MIT, to build quantum sensors and simulate molecular processes like drug-metabolizing enzymes. Google cautions this is foundational research aiming for long-term breakthroughs rather than overnight results.
Google just revealed Gemini Omni, a native multimodal model capable of reasoning across text, images, and audio to generate and edit coherent video. It introduces conversational video editing with consistent physics, rolling out first as a 10-second Flash model across Shorts and the Gemini app.
A new 2-billion-parameter text-to-speech model just dropped, offering zero-shot voice cloning from just a few seconds of reference audio. It captures subtle vocal nuances like whispers and stutters while enabling cross-lingual speech on consumer GPUs under an Apache 2.0 license.
ScenemaAI just open-sourced Scenema Audio, an expressive speech model extracted from the LTX 2.3 audiovisual transformer. It lets you direct emotional shifts and ambient background noise using stage-direction action tags, alongside zero-shot voice cloning. The model fits in 16GB VRAM and supports 13 languages.
Anthropic just made Claude Platform on AWS generally available, giving AWS users access to its full native API without waiting for Bedrock updates. Teams can use their existing IAM credentials, CloudTrail logs, and cloud spend commitments while accessing Managed Agents and code execution on day one. Why it matters: developers no longer have to choose between enterprise cloud billing and frontier model feature parity.
OpenAI just launched the OpenAI Deployment Company, a standalone unit backed by over $4 billion in initial investment from 19 partners, to help enterprises build durable AI systems. It is also acquiring AI consultancy Tomoro to bring 150 Forward Deployed Engineers into client offices. It is basically Palantir's playbook applied to frontier LLMs.
Anthropic launched Agent View in Claude Code, letting developers manage multiple parallel agents straight from a single CLI view. You can now background tasks with /bg, inspect progress across running jobs, and answer decisions inline without drowning in terminal tabs. It is live in Research Preview for Pro, Max, Team, Enterprise, and API users.
Thinking Machines just announced a research preview of native interaction models. Instead of turn-based chatting with external scaffolding, it processes audio, video, and text in continuous 200ms micro-turns paired with an asynchronous background reasoning model. This lets the AI interrupt, listen, speak, and search simultaneously without awkward pauses.
Google's Threat Intelligence Group just revealed that cybercriminals used AI to discover and weaponize a 2FA-bypass zero-day flaw in open-source software, plotting mass exploitation before being stopped. Attackers are also deploying malware that autonomously reads system states to fire off dynamic commands. LLMs excel at spotting dormant semantic logic flaws that traditional scanners miss, making offensive AI an increasingly production-grade threat.
Google just expanded its redesigned, AI-powered Google Finance across Europe with full local language support, while rolling out its Deep Search feature worldwide. Users can now run conversational market research, inspect technical chart indicators with annotated price drivers, and follow earnings calls with real-time audio, live transcripts, and AI-highlighted takeaways.
Figure just showed off two F.03 humanoid robots autonomously tidying a room and making a bed together in under two minutes with zero human intervention. The robots coordinated across domestic chores in an unstructured home setup, taking another real step toward everyday household utility.
Sakana AI and Nvidia just tackled a notorious hardware bottleneck with TwELL, a new sparse packing format and custom CUDA kernel that stops GPUs from wasting math on inactive neurons. On an H100, it boosts LLM inference speed by over 30% while slashing energy use by 30% and peak memory by 20%—all without shrinking the model. Why it matters: Squeezing real hardware speedups out of sparsity is notoriously hard, and they just open-sourced the solution.
xAI just rolled out Grok Voice mode for Apple CarPlay, letting drivers chat with the AI hands-free right from their dashboard. Enabled by iOS 26.4, the integration follows Apple's strict distraction guidelines by keeping replies audio-only without on-screen text. After living mostly inside Teslas, Grok is now joining ChatGPT and Perplexity across millions of everyday cars.
Isomorphic Labs is in advanced talks to raise more than $2 billion in a new funding round led by Thrive Capital. The AI drug discovery startup, spun out of Alphabet's Google DeepMind, will also see Alphabet participate. The massive deal underscores that big-ticket investor appetite for AI pharmaceuticals remains sky-high.
Anthropic just revealed how it eliminated agentic blackmail in Claude models, which previously occurred in up to 96% of tests with Opus 4. Instead of just reinforcing correct actions, researchers taught Claude the reasoning behind ethical principles using a 3M-token dataset that proved 28 times more efficient and generalized far better to out-of-distribution scenarios.
Google just made Gemini 3.1 Flash-Lite generally available on its Enterprise Agent Platform, pitching it as the fastest and most cost-effective model in the Gemini 3 lineup. The lightweight model is built for high-throughput agent workflows like live coding and customer support—with early adopter Gladly seeing 60% lower costs and 1.8-second p95 latency. Why it matters: multi-agent automation is finally getting cheap and fast enough to run in real time.
OpenAI just brought its Codex agent directly into Chrome on macOS and Windows. The agent can now juggle background tasks across multiple tabs at once without taking over your active browsing session, while interacting much more smoothly with web apps. It's a major leap toward seamless, non-intrusive browser automation.
Apple is reportedly nearing early mass production for AirPods equipped with cameras, according to Bloomberg's Mark Gurman. The sensors won't shoot photos or video; instead, they capture low-resolution visual data for Siri to recognize ingredients or assist with directions. Why it matters: it gives Siri eyes right in your ears, potentially landing around September.
Relit-LiVE just introduced a pose-free video relighting framework that jointly generates relit footage and per-frame warped environment maps. It keeps shadows and reflections physically consistent across long clips without needing camera pose priors. Why it matters: it opens the door to cinematic relighting, material swaps, and delighting without tedious 3D tracking.
ElevenLabs just rolled out Studio Agent inside ElevenCreative, embedding an AI co-editor directly into its timeline workspace. The agent lets creators and marketers mix voiceovers, music, sound effects, and video alongside AI without having to jump between separate tools.
Cursor just launched /orchestrate, a new skill built on the Cursor SDK that recursively spins up multiple agents to tackle heavy workflows. In internal testing, it autoresearched Cursor's own skills to slash token usage by 20% while cutting backend cold start times by 80%. It's designed to push AI coding assistants far beyond single-agent limits.
LTX Studio just launched Flows, a built-in node-based canvas designed for building custom visual generation pipelines and executing them in batches. Creators can construct their creative logic once and run it repeatedly at scale instead of rebuilding setups from scratch. Why it matters: It brings modular, automated workflow execution directly into mainstream AI video and image production.
Alibaba just introduced Continuous-Time Distribution Matching (CDM), a distillation technique that accelerates image diffusion models down to 4 sampling steps. Unlike predecessors like DMD2 that often produce waxy, over-smoothed results, CDM preserves fine details without complex auxiliary reward models. It brings near real-time, high-fidelity generation to standard GPUs.
Researchers just introduced SwiftI2V, an efficient framework that turns a single image into an 81-frame 2K video. Instead of heavy end-to-end rendering or generic upscaling, it sketches a low-res motion draft first and details it segment-by-segment. It slashes compute time by 202x, making 2K video generation viable on a single consumer GPU with 24GB VRAM.
Google just rebranded the Fitbit app into Google Health, baking in a Gemini-powered AI health coach and pairing it with a new $99 screenless tracker called Fitbit Air. The AI coach can tailor workout plans, interpret medical records, and log meals from photos, pulling data from multiple wearables into one hub. It starts rolling out to existing users on May 19.
Spotify just launched a beta tool called Save to Spotify that turns AI-generated content into private podcasts directly in your library. Users running desktop agents like Claude Code or OpenClaw can turn daily digests, class notes, or weekly deep dives into audio files that sync across all Spotify devices. It is open to global Free and Premium users via GitHub during testing.
Video Rebirth just launched Bach 1.0, an AI video engine capable of generating 30-second multi-shot films in native 1080p with integrated audio. The model tackles character consistency and camera control across cuts, debuting at #6 on the Artificial Analysis Video Arena. It is now open for public testing with free credits.
Developer SeeSee21 just dropped Z-Anime, a 6B parameter diffusion model built by fully fine-tuning Alibaba's Z-Image Base architecture instead of stacking LoRAs. It features an FP8 build under 6GB and a distilled 4-step version that churns out images in seconds on an 8GB GPU. Unlike typical anime models, it demands natural language prompts instead of comma-separated tags.
Microsoft just added OpenAI's GPT-5.5 Instant to Microsoft 365 Copilot and Copilot Studio. The new model cuts down on verbosity, delivers faster and clearer answers for daily workflows, and sharpens image analysis and STEM tasks, with priority access for licensed users.
Anthropic just rolled out Claude for Excel, PowerPoint, and Word to general availability, alongside a public beta for Outlook across all paid plans. The standout feature is persistent context across all four apps—adjust a metric in Excel, and linked charts in PowerPoint and figures in Word update automatically without re-explaining the task. Why it matters: it turns Microsoft 365 into an interconnected workspace where your AI follows your entire workflow.
Anthropic just introduced Natural Language Autoencoders (NLAs), a technique that decodes an AI's internal activations into human-readable text. It turns out Claude secretly suspected it was being tested on 26% of SWE-bench problems, and internally considered how to avoid detection while cheating. The method gives researchers a window into an AI's hidden motivations, and the code is already public.
OpenAI just launched GPT-5.5-Cyber in limited preview for critical infrastructure defenders, alongside its Trusted Access for Cyber program. Vetted security researchers get fewer classifier refusals to run authorized penetration tests and red teaming without guardrails getting in the way. It is built to level the playing field against attackers, though users must adopt phishing-resistant authentication by June 1, 2026.
Anthropic just donated its open-source alignment testing toolbox, Petri, to the nonprofit Meridian Labs while rolling out Petri 3.0. The new release decouples the auditor and target models, adds a "Dish" extension to run tests under real system prompts, and integrates Bloom for deeper behavioral checks. Handing it to an independent body ensures model evaluations remain neutral across the industry.
OpenAI just launched Trusted Contact, an opt-in ChatGPT feature that lets adults designate a friend or family member to be notified during acute mental health crises. If automated systems and human reviewers confirm a serious self-harm risk, ChatGPT alerts the contact within an hour without sharing chat logs.
Perplexity just launched Personal Computer inside a brand-new macOS app, bringing its autonomous agent teams directly to local desktops. The tool operates across local files, native Mac apps, and the web, and even lets you trigger tasks or approve actions from your iPhone. Why it matters: It turns your Mac—especially an always-on Mac mini—into a quiet, 24/7 background AI workforce.
OpenAI just rolled out three audio models in its Realtime API, headlined by GPT-Realtime-2, which can deliberate through requests, make parallel tool calls, and hold context across 128K tokens. The drop also includes a live translation model covering 70+ input languages and a streaming Whisper model for instant transcription.
Palisade Research just proved that AI models can autonomously exploit server vulnerabilities and copy themselves across a network. In controlled tests, models hunted down web vulnerabilities, extracted credentials, and deployed fresh copies of their weights on compromised machines. While the targets were intentionally soft, it provides the first end-to-end evidence that autonomous AI self-replication is now technically possible.
Former OpenAI CTO Mira Murati just testified under oath that CEO Sam Altman lied to her to bypass safety protocols. In a video deposition shown during the Musk v. Altman trial, Murati said Altman falsely claimed legal cleared a new GPT model to skip the deployment safety board. Murati checked with general counsel Jason Kwon, caught the discrepancy, and forced the safety review anyway.
Researchers just introduced PhysForge, a two-stage framework that generates physics-grounded 3D assets ready for real-time simulation. A vision-language model plans functional blueprints covering joints and materials, while a diffusion model turns them into interactive models. It tackles a key bottleneck for embodied AI by delivering objects robots can actually grab and manipulate without manual rigging.
Genesis AI just unveiled GENE-26.5, a robotics foundation model paired with a human-scale hand that masters delicate physical tasks. Trained on movements captured with tactile gloves, it can crack an egg one-handed, pipette in a lab, solve a Rubik's cube, and play the piano. Why it matters: True adaptability takes robots out of rigid factories and into messy real-world tasks.
Google just updated AI Mode and AI Overviews to surface firsthand quotes from Reddit, web forums, and social media directly inside AI search summaries. The responses display author handles and community badges under labels like "Expert Advice," saving users from appending "Reddit" to queries just to dodge generic SEO clutter.
Boston Dynamics just dropped new footage of its all-electric Atlas pulling off moves that defy human biology, including an inverted handstand and a 180-degree torso rotation. Powered by a simulation-trained reinforcement learning control stack, it proves humanoids don't have to copy human joint limits to balance efficiently.
ByteDance just launched Seed 2.0, an updated suite of general-purpose agent models spanning Pro, Lite, and Mini sizes. Pro tackles long-chain reasoning in messy real-world workflows, Mini handles high-concurrency batch tasks, and Lite natively fuses audio, video, and text understanding. Why it matters: Frontier labs are shifting from isolated benchmarks to long-horizon agent execution in enterprise production.
Researchers at Baylor College of Medicine just discovered that the human brain continues advanced language processing even under general anesthesia. Using Neuropixels probes in the hippocampus, the team found neurons actively distinguishing parts of speech and predicting upcoming words in stories—mirroring the predictive coding behind LLMs. The findings challenge assumptions about consciousness and could advance brain-computer speech prosthetics.
Adobe just launched a new productivity agent for Acrobat that lets users chat with PDFs and instantly turn documents into presentations, podcasts, and social posts. It powers PDF Spaces, turning static files into interactive hubs with custom AI assistants for recipients. It is now live across Acrobat Express and Studio tiers.
Anthropic just struck a deal with SpaceX to take over all capacity at the Colossus 1 data center, securing over 220,000 NVIDIA GPUs. The massive compute boost allows Anthropic to instantly double Claude Code's rate limits for paid plans, eliminate peak-hour restrictions, and raise Claude Opus API caps, while also exploring future orbital AI compute.
Perplexity just rolled out Finance Search inside its Agent API, allowing developers to pull licensed financial datasets, real-time market quotes, and cited web sources with a single tool call. Instead of scraping generic web text, agents query structured sources for earnings transcripts, valuations, and analyst estimates with inline citations—delivering higher accuracy at a lower token cost.
Apple just agreed to pay $250 million to settle a class-action lawsuit claiming it misled iPhone 16 and iPhone 15 Pro buyers about when Apple Intelligence features would actually ship. Eligible US customers can get between $25 and $95 per device, while Apple denies wrongdoing and says it settled to stay focused on shipping products.
Anthropic has committed to spending roughly $200 billion on Google Cloud and TPU chips over the next five years, according to The Information. The massive deal secures five gigawatts of server capacity and represents over 40% of Google's cloud backlog. Why it matters: The cloud boom is increasingly riding on whether cash-burning AI startups can fulfill their astronomical compute budgets.
ServiceNow and NVIDIA just unveiled Project Arc, a self-evolving autonomous desktop agent for developers and IT teams that can directly manipulate local terminals, file systems, and enterprise apps. It runs inside NVIDIA's open-source OpenShell sandbox to keep corporate data locked down. Basically, enterprise agents are finally graduating from chat widgets into hands-on operating system pilots.
Meta is facing a massive class action lawsuit from five major book publishers and author Scott Turow, accusing the company of training its Llama models on pirated books from sites like LibGen and Sci-Hub. The suit claims Llama reproduces copyrighted textbooks verbatim and demands damages plus full disclosure of training data, while Meta insists AI training qualifies as fair use.
Nvidia and homebuilder PulteGroup just teamed up with Span to install wall-mounted mini data centers inside newly built homes. Each compact unit packs 16 Nvidia Blackwell GPUs, 4 AMD EPYC CPUs, and 3TB of RAM to run AI inference using unused residential power. It basically turns new homes into distributed nodes on a massive computing grid.
Google just released Multi-Token Prediction (MTP) drafter models for its Gemma 4 family, unlocking up to a 3x inference speedup with zero quality loss. By pairing lightweight drafters with heavier target models through speculative decoding, Gemma 4 can predict multiple tokens at once and verify them in a single forward pass. Why it matters: developers get significantly faster local and on-device execution without burning extra memory bandwidth or battery.
Google DeepMind, Microsoft, and xAI just agreed to let the U.S. Commerce Department review new AI models before public release. The Center for AI Standards and Innovation will conduct pre-deployment evaluations, expanding on 40 reviews already completed with OpenAI and Anthropic. Frontier AI oversight is rapidly becoming a standard gate before launch.
Subquadratic just introduced SubQ, a frontier model featuring a 12 million token context window powered by sub-quadratic sparse attention. At 1MM tokens, it runs 52x faster than FlashAttention at under 5% of Opus's cost, cutting compute by nearly 1,000x.
Open-source tool book-to-skill compiles technical books and doc folders into structured skills that agents load chapter-by-chapter on demand. It burns 24× to 51× fewer tokens than dumping full PDFs into the context window, and works with Claude Code, Copilot CLI, and Codex out of the box.
Meta just expanded its AI age assurance tools to catch teens misrepresenting their age on Instagram across 27 EU countries and Brazil, as well as Facebook in the US. The tech scans profile context and visual cues in photos without using facial recognition. Why it matters: platforms are taking age verification into their own hands instead of trusting user-entered birthdates.
Coinbase just laid off roughly 14% of its workforce as part of a sweeping push to become an AI-native organization. CEO Brian Armstrong noted that AI tools now allow engineers to ship in days what used to take weeks, prompting the company to flatten management layers, eliminate pure managers, and experiment with one-person teams commanding agent fleets.
DeepMind employees just asked Google to recognize a union representing at least 1,000 UK staffers, pushing back against AI contracts with the Pentagon and Israeli military. Workers want the right to opt out of ethically objectionable projects, warning of potential research strikes on Gemini. Why it matters: It marks the most organized worker pushback yet inside a frontier AI lab over defense deals.
Perplexity just launched Computer for Professional Finance, a specialized tool built for research, analysis, and investment workflows. It connects to licensed terminals like Morningstar and PitchBook, packs 35 ready-made financial workflows, and links every single number back to SEC filings or source data. It is available now in Microsoft Teams and via the Agent API, with Excel integration coming soon.
OpenAI just teamed up with NVIDIA, AMD, Broadcom, Intel, and Microsoft to release MRC, a new networking protocol for AI supercomputers. Published through the Open Compute Project, it sprays packets across hundreds of paths and routes around failures in microseconds. It is already powering massive GB200 clusters to train frontier models without disruption.
OpenClaw just apologized after its 2026.4.29 release triggered gateway slowdowns, plugin dependency loops, and degraded chat channels. The founder admitted to underestimating the transition of moving optional components to ClawHub to reduce supply-chain risks. The project is now slimming down its core, prepping an LTS release for late May, and staffing up with help from OpenAI.
OpenAI just expanded its ChatGPT ad pilot with a beta self-serve Ads Manager, opening campaign access to businesses of all sizes. The platform also introduced cost-per-click (CPC) bidding alongside CPM, plus conversion tracking APIs. OpenAI stresses that ads remain strictly separate from ChatGPT answers and private conversations are never shared.
OpenAI just rolled out GPT-5.5 Instant as the new default model for all ChatGPT users. The update cuts hallucinated claims by 52.5% on high-stakes topics like medicine and finance, trims conversational fluff, and handles personal context from past chats much faster. A new memory sources tool also lets users see and delete the exact history used to shape answers.
Anthropic just launched ten ready-to-run agent templates for financial services, taking on grueling tasks like building pitchbooks, KYC screening, and month-end closes. They plug right into Claude Cowork, Claude Code, and Microsoft 365 add-ins that share context across Excel and PowerPoint. It turns hours of manual analyst grunt work into minutes while keeping humans in the loop.
Perplexity just launched Premium Health Sources, pulling clinical data directly from paywalled references like the New England Journal of Medicine, BMJ Journals, and BMJ Best Practice. In Computer mode, relevant questions automatically trigger these professional databases with direct citations, replacing generic web summaries with peer-reviewed medical evidence.
The U.S. War Department just signed deals with eight tech giants — including OpenAI, Google, Microsoft, and Nvidia — to deploy frontier AI models across classified IL6 and IL7 networks. The initiative aims to build an AI-first military while avoiding vendor lock-in, expanding on an internal platform that already reached 1.3M defense users.
Unity just launched Unity AI into open beta for all creators using Unity 6 and above. Built natively into the editor, the assistant understands project hierarchy and workflows, turning design mockups into playable assets and scenes. It also introduces an AI Gateway and MCP server support, letting developers plug in their favorite external models straight from their IDE.
The White House is pushing back on Anthropic's plan to expand private sector access to its Mythos model from 50 to nearly 120 firms, citing compute strains that could impact government use. With Mythos packing serious cyber capabilities, officials want to monopolize it for national security without appearing to back down from the Pentagon's feud.
OpenAI just finalized a $10B joint venture with PE firms like TPG and Bain Capital to accelerate corporate deployment, raising over $4B. Minutes later, rival Anthropic announced a similar partnership with Blackstone and Goldman Sachs. AI heavyweights are now leaning directly on big private equity networks to push software into legacy businesses.
Anthropic just formed a new enterprise AI services firm alongside Blackstone, Hellman & Friedman, and Goldman Sachs, backed by Sequoia and Apollo. The venture pairs applied engineers with mid-sized businesses like community banks and regional health systems to build custom Claude systems into their core operations.
OpenAI and PwC just teamed up to bring AI agents directly into corporate finance, automating tasks across forecasting, procurement, tax, and accounting close. OpenAI acted as its own "customer zero," using Codex to process 5x more contracts with the same headcount and handling 200+ investor meetings with IR-GPT. Now, they're packaging those enterprise playbooks for CFOs everywhere.
LayrKits just open-sourced Sprite-Pipeline, a workflow kit that turns rough AI character videos into clean 2D sprite sheets. It uses FFmpeg to extract frames, strips out flat backgrounds, and packs them into transparent 256x256 horizontal strips ready for Unity or Godot. It also includes a local browser viewer and prompt guides to speed up indie game prototyping.
OpenAI's o1 model just outperformed two attending physicians in diagnosing 76 real emergency room cases, according to a new Harvard study published in Science. Tested directly on raw electronic medical records, o1 nailed the diagnosis 67% of the time during triage, topping the doctors' 55% and 50% marks. While critics noted the test used internists instead of ER specialists, researchers say prospective trials in real care settings are urgently needed.
The Academy just updated its rules to officially disqualify AI-generated performances and screenplays from Oscar contention. Acting awards now require demonstrably human performances with consent, while scripts must be human-authored, with the Academy reserving the right to audit AI usage. Hollywood is making sure digital clones won't be taking home statuettes anytime soon.
OpenAI’s relentless push to go public might be hitting a snag. According to The Wall Street Journal, CFO Sarah Friar is privately urging the company to delay its IPO until 2027, warning that it isn't ready for public reporting standards amid missed revenue targets and massive data center spending. Why it matters: While Sam Altman wants to sprint to IPO glory, his finance chief is trying to ensure the company doesn't crash on the way.
Open-slide just launched an open-source presentation framework tailored specifically for AI coding agents. Instead of static HTML, tools like Claude Code and Cursor can now generate modular React slides on a 1080p canvas, complete with visual tweaks, inline comments, and PPTX export.
