What is actually shipping in AI coding right now. New models, agent updates, the MCP ecosystem and the business moves behind them. Every item is a short summary in our own words with a link to the original, so you can scan the field in a minute and dig in where it counts.
Build 2.1.183 stops auto mode from quietly running hard resets, checkout discards, clean -fd or stash drops unless you actually asked to throw work away. There is also a new mouse-wheel scroll setting and a pile of billing and performance fixes.
JetBrains pushed Junie out of beta and rebuilt it on the Agent Communication Protocol. The headline trick is agentic debugging: the agent drives the actual IDE debugger instead of guessing, and you can now point it at a whole directory of guideline files.
Cursor's June update leans hard into background work: always-on agents that chew through repetitive tasks, an /automate skill that builds workflows from plain language, and expanded GitHub and Slack triggers. Cloud agents can now drive a browser to produce demos.
Cognition folded Windsurf into Devin Desktop on June 2, turning the editor into a hub for managing autonomous agents. An Agent Command Center gives you a Kanban view of active sessions, and Devin now runs as a sandboxed background agent you dispatch from inside the IDE.
Fable 5 went GA on June 9, opening Anthropic's new Mythos-class tier to everyday use with extra safeguards. On the coding benchmarks it sits at the top of the pile, and Claude Code running on Fable 5 posts the strongest terminal scores Anthropic has published.
The Model Context Protocol's 2026-07-28 spec is now a release candidate and it is the biggest revision since launch. A stateless core lets servers scale on plain HTTP, an Extensions framework adds Tasks and server-rendered MCP Apps, and authorization gets aligned with OAuth and OpenID Connect.
The Model Context Protocol registry counted roughly 9,650 latest server records and nearly 29,000 versioned records in late May. Monthly SDK downloads have blown past 97 million, a staggering climb from around 2 million when the protocol launched.
As of June 1, Copilot moved every plan onto usage-based billing through GitHub AI Credits, and code review now also eats Actions minutes. A new Copilot Max plan lands at 100 dollars a month with a 20,000-unit budget aimed at people running sustained agent workflows.
Codex's June update extends Computer Use and remote control to Windows, broadens its profile and thread tooling, and sharpens search and keyboard shortcuts. Chrome context capture got faster too. It builds on the GPT-5.x line now powering the agent.
OpenAI moved Codex onto GPT-5.5, with NVIDIA detailing the infrastructure behind the agent push. Codex crossed two million weekly users earlier in the year and OpenAI keeps framing it less as a coding tool and more as a general enterprise agent platform.
Google made Gemini 3 Flash generally available as its speed-and-cost tier for agentic and coding tasks, sitting under the heavier 3.1 Pro. Cursor, GitHub, JetBrains, Replit and others have already wired the Gemini 3 line into their tools.
Google's 3.1 Pro arrived earlier in the year claiming roughly double the reasoning of 3 Pro, a 1M-token context window and 65K-token output. It tops a majority of the benchmarks Google tracks, and the third-party tools picked it up quickly.
Anthropic's June Claude Design update is rolling out in beta to paid users with imported design systems and a direct visual canvas you can edit. The part builders care about: a handoff that turns an AI-generated prototype into working code in Claude Code.
Cognition is extending Devin Desktop with custom background agents whose context reaches every engineer's laptop, so humans and agents share the same picture instead of starting cold. NVIDIA has joined the research preview for the multi-agent setup.
A mid-June scoring run put Claude Fable 5 near the top of SWE-bench Verified at 95 percent, with Opus 4.8 leading the tougher vendor SWE-bench Pro numbers and GPT-5.4 setting the pace on the standardized split. Terminal-Bench v2 got measured alongside.
GitHub gave Copilot code review more knobs: organization runner controls, content exclusion support, and the removal of the character cap on repository custom instructions. Reviews run on GitHub Actions, with self-hosted or large runners available for heavier jobs.
Alongside the IDE agent, JetBrains shipped a standalone Junie CLI that is model-agnostic and built for terminals, CI/CD pipelines and GitHub or GitLab workflows. It is the piece that lets Junie run where there is no editor open.
Musk used a launch event to show off Grok Imagine 1.5, the biggest jump yet for xAI's generative media model. The headline is that it no longer stops at stills: the model now produces full video clips, pushing xAI directly into the Veo and Kling fight.
Alibaba added another entry to its Qwen coding line with Qwen3-Coder-Next, keeping the open-weight family moving at the same fast cadence the rest of the field has settled into this year.
Groq put llama-3.1-8b-instant and llama-3.3-70b-versatile on its deprecation list. If you have hard-coded either model id into a Groq integration, now is the moment to swap to a supported one before they go dark.
Google moved gemini-3.1-flash-image (Nano Banana 2) and gemini-3-pro-image (Nano Banana Pro) into general availability and folded in video-to-image generation, putting both image tiers on the stable API for production use.
xAI made Grok 4.3 generally available through AWS Bedrock, carrying a 1M-token context window and configurable reasoning so enterprise teams can dial effort up or down per call.
xAI shipped a free Microsoft 365 add-in that drops Grok straight into Word for drafting and rewriting, with the model able to pull in live web results and data from X while you write.
Salesforce planted a flag in the Model Context Protocol ecosystem, shipping DX, Heroku and MuleSoft MCP servers around its Agentforce agent platform. It is another large vendor treating MCP as the default wiring for agents.
Anthropic donated MCP to a new Agentic AI Foundation under the Linux Foundation, with OpenAI, AWS, Google and Microsoft signing on as backers. Vendor-neutral governance is a strong signal that MCP is becoming shared infrastructure rather than one company's spec.
OpenAI introduced GPT-5.5, framing it as its best model yet for coding, research and multi-tool task completion. It is the foundation the company is now building its agent products on top of.
OpenAI promoted GPT-5.5 Instant to the default model for everyone on free ChatGPT, retiring GPT-5.3 Instant. Casual users get a meaningful capability bump without changing a single setting.
Anthropic released a new flagship that tops its intelligence benchmarks and added a dynamic workflows tool to Claude Code aimed at very large problems, plus per-task effort controls so you can trade speed against depth.
Coverage of the Opus 4.8 release zeroed in on the new dynamic workflows feature in Claude Code for very large-scale problems, paired with a cheaper fast mode for everyday work.
DeepSeek put out two open-weight preview models, a 1.6T-parameter V4-Pro and a 284B V4-Flash, both with a 1M-token context window. As of June they are still in preview, but the open weights make them worth watching.
Google unveiled Gemini 3.1 Pro, a top-tier model with strong reasoning, multimodal and coding benchmark numbers. It sits at the head of the Gemini 3 line that third-party tools rushed to adopt.
Black Forest Labs launched FLUX.2, a 32B-parameter lineup spanning a commercial Pro API, an open-weight dev model and a sub-second klein variant. The split lets you pick speed, control or licensing without leaving the family.
Midjourney shipped V8 with roughly five times faster generation, reliable text rendering and better prompt adherence, followed quickly by a V8.1. Text in images has long been the weak spot, so this is the upgrade users were waiting on.
OpenAI shut down the standalone Sora web app and folded video generation into ChatGPT, with the Sora API slated to sunset in September. If you built on the Sora API, start planning the migration now.
Kling 3.0 added a multi-shot storyboard mode and native audio sync that holds across cuts, putting it shoulder to shoulder with Veo 3.1 on cinematic realism. Storyboarding inside one model is the kind of feature short-film creators actually use.
Veo 3.1 set the pace for narrative AI video on prompt adherence, native audio and 4K resolution. For anyone trying to get a clip that matches the brief on the first try, that combination is the one to beat.
Seedance 2.0 entered testing and quickly turned into the most-poked-at new AI video model among creators, a sign ByteDance is a serious contender in a field crowded with Veo, Kling and Runway.
ElevenLabs released Music v2 with genre-shifting and section-by-section composition, then trimmed Music pricing across its tiers. More control plus cheaper credits is an aggressive move into Suno and Udio territory.
Stability AI launched Stable Audio 3.0, a four-model family with open weights for three of the variants and tracks running up to roughly six and a half minutes. The open weights make it the obvious base for self-hosted music tooling.
Meshy released version 6 with a big jump in mesh fidelity, closing more of the gap between AI-generated 3D and hand-sculpted assets. Higher base quality means less cleanup before a model is game-ready.
Meshy launched its Meshy Labs AI incubator at GDC alongside an AI-native game and said it had tripled annual recurring revenue to $30M. The numbers show real money is now flowing into generative 3D, not just demos.
OpenAI raised the largest private venture round on record at $122B, pushing its valuation to around $852B, with Amazon named as the exclusive third-party cloud partner. The scale of capital flowing into frontier labs keeps setting new highs.
Anthropic closed a roughly $65B Series H that took its reported post-money valuation to about $965B. The raise plants it at the very top of the private AI market alongside OpenAI.
Global venture funding hit a record of about $300B in the first quarter, driven mostly by enormous frontier-lab AI rounds. The concentration of capital in a handful of labs is reshaping the whole startup market.
NVIDIA released new open Isaac GR00T robot models that handle natural-language multistep tasks, plus Cosmos world models for generating synthetic training data. It is more groundwork for general-purpose robotics built on simulation.
Sony AI revealed Project Ace, a real-world autonomous robotics system that can hold its own against elite human table-tennis players. Pulling that off in the physical world, not a simulator, is the hard part.
Researchers at Ames Laboratory built DuctGPT, a physics-trained AI workflow that designs new permanent-magnet materials without rare-earth elements. It is a concrete case of AI doing materials discovery with real supply-chain stakes.
SpaceX has struck a deal to purchase Cursor, the company behind the popular AI-assisted code editor, for roughly $60 billion in stock. The agreement closes out talks that first surfaced earlier in the year and folds one of the best known vibe-coding tools into Elon Musk's rocket firm.
xAI introduced Grok Build, a command-line coding agent built for serious software work and complex engineering tasks. The tool is in early beta and, at launch, is limited to SuperGrok Heavy subscribers paying $300 a month.
OpenAI rolled out a Codex update that gives developers multi-purpose agents able to operate across a wider range of tasks and act more proactively, starting with computer use. The release reads as early scaffolding for the all-in-one app OpenAI is reportedly preparing.
Anthropic has opened up Claude Cowork, its assistant for handling everyday tasks on your computer, to all Pro subscribers paying $20 a month. The feature was previously restricted, and the move puts agentic desktop help in front of a much wider audience.
OpenAI released Prism, an agentic app modeled on the Claude Code workflow but aimed at scientific work rather than general software. It pitches researchers a terminal-like assistant that can carry out multi-step tasks across their projects.
LinkedIn is letting members display proficiency with AI coding tools directly on their profiles, leaning into the rise of vibe coding. The feature launches with partners including Replit, Lovable, Descript and Relay.app.
At Build 2026, Microsoft announced Project Solara, a platform designed around AI agents rather than traditional apps. The company showed it powering concept hardware including a smart display and a smart key badge, framing agents as the next big platform shift.
Anthropic launched Fable, a model that brings the abilities of its unreleased Mythos system to the public. In the company's own benchmarks, Fable beat its prior flagship Opus 4.8 as well as competing models from OpenAI and Google.
OpenAI released GPT-5.2 as its answer to the latest models from Google and Anthropic. The update positions the company to keep pace in a fast-moving race where rivals have been trading the lead on reasoning benchmarks.
Google began rolling out Gemini 3 Flash, a more efficient version of its latest model that delivers pro-level reasoning at a fraction of the flagship's cost. On some benchmarks it comes out ahead of OpenAI's GPT-5.2.
OpenAI updated ChatGPT's image generation to run about four times faster than before while following instructions more closely. The improvements show up especially when users ask for tweaks to an already generated picture.
Google added Personal Intelligence to its image tools, letting Gemini pull from Gmail, Search and YouTube to tailor the pictures it creates. The idea is to make personalized image generation faster by drawing on what Google already knows about you.
OpenAI signed a multi-year agreement with Getty Images that will bring Getty's licensed content into ChatGPT and OpenAI search results. The partnership gives the chatbot a stock-image source with clearer rights behind it.
OpenAI is shutting down its dedicated Sora video app, telling users it's saying goodbye to the standalone product. The company plans to fold its Sora video generation model directly into ChatGPT instead.
According to reports, OpenAI intends to add Sora's video generation directly to ChatGPT rather than keep it in a separate app. The shift would put text-to-video tools alongside the chatbot's existing image and coding features.
Warner Music Group dropped its legal case against AI music platform Suno in exchange for a licensing deal covering its artists' songs and likenesses. Suno also said it is rolling out newer, licensed models in 2026 and will retire its current ones.
An Atlantic investigation published searchable databases revealing the scale of music used to train AI systems, including one set with 12 million tracks and another with 9 million. The findings add fuel to ongoing fights over whether platforms like Suno and Udio can claim fair use.
At Unreal Fest, Epic detailed an experimental MCP plugin that lets developers connect models like Claude and Gemini to Unreal Engine. The integration can reach core systems such as blueprints, assets and materials to automate creation, testing and optimization, and Epic wants it baked into UE6.
Roughly 20 percent of the demos in the latest Steam Next Fest included a generative AI disclosure, reflecting how common the tools have become in game development. The figure offers a concrete read on how quickly studios are folding AI into their pipelines.
Boston Dynamics unveiled a production-ready version of its all-electric Atlas robot at CES 2026. Hyundai plans to put the humanoid to work in its car plants starting in 2028, handling parts sequencing before moving on to assembly tasks.
Microsoft's latest Visual Studio release pushes AI deeper into the editor, including a Profiler Agent that helps developers track down and fix performance problems without being profiling experts. The update positions AI assistance as a default part of the everyday coding workflow rather than an add-on.
OpenAI released a dedicated Codex application that lets developers coordinate multiple AI coding agents across projects, moving past simple chat-driven code generation. The launch arrives as companies debate how much autonomy to grant these tools and how to govern them.
JetBrains unveiled JetBrains Central, an agentic development platform that gives teams oversight and management across their AI coding agents. It works alongside JetBrains Air and the model-agnostic Junie agent, with an early access program slated to begin in the second quarter of 2026.
Google is consolidating its developer AI tools under Antigravity, aiming to support the full agentic software lifecycle instead of offering disconnected assistants. The idea is a persistent layer where project context, run history, and agent state carry across coding, testing, debugging, and deployment.
InfoWorld lays out the developments it expects to define the year, including self-verifying agents that catch and correct their own mistakes mid-task. The piece frames error accumulation in multi-step workflows as the main barrier that needs solving before agents can scale in the enterprise.
The case for compact models is growing as teams weigh speed, cost, privacy, and lower resource demands against the heavyweight LLM approach. InfoWorld argues this shift is changing how architects plan AI systems rather than just trimming a few parameters.
InfoWorld surveys specialized models tuned for narrow fields, from medicine to law to finance, including Microsoft's PubMed-trained BioGPT and JPMorgan Chase's contract analysis system. The takeaway is that domain-specific tuning is becoming a serious alternative to relying on one general model for everything.
This guide breaks down the measurements that matter when evaluating large language models, covering accuracy, cost, latency, and safety dimensions. It is aimed at teams trying to compare models with more rigor than vibes and benchmarks alone.
A new technique predicts several tokens at once and claims a roughly threefold speedup in inference without needing separate draft models. If it holds up, it could cut serving costs for production systems that currently lean on speculative decoding.
Google's Gemini 2.5 Flash Image model brings faster generation and editing through the Gemini API, AI Studio, and Vertex AI for enterprise use. It targets developers who want to wire image creation and manipulation directly into their applications.
Microsoft's Phi-4-multimodal is a 5.6 billion parameter model that uses a mixture-of-LoRAs approach to handle speech, vision, and language together. The company pitches it as an efficient, scalable option for developers building multimodal features.
Google released tooling that lets developers run agentic workflows on their own machines using the 12-billion-parameter Gemma 4 12B from DeepMind. The push is about keeping agent execution local rather than routing everything through paid cloud APIs.
DiffusionGemma is a 26B mixture-of-experts model built on the Gemma 4 family and Google's Gemini Diffusion research, generating text without strict sequential decoding. The design aims to boost text output throughput by breaking away from token-by-token generation.
Google's Gemma 4 family spans several sizes and quantizations so it can run on everything from servers down to commodity PCs. InfoWorld's review highlights its reasoning, tool use, and multimodal vision and audio support as standout features for local deployment.
InfoWorld digs into world models that learn the rules of reality from data and can run interactive, game-engine-free simulations. It points to examples like Decart's playable AI environments and World Labs' Marble, which rebuilds 3D scenes from still images and lets users reshape them on the fly.
InfoWorld rounds up Model Context Protocol servers that give AI agents real abilities across Git, CI/CD, infrastructure as code, observability, and documentation. The piece reflects how quickly MCP has become a connective standard for agent-driven operations.
At its Data and AI Summit, Databricks introduced Genie ZeroOps, an agentic capability that automates monitoring, investigation, and remediation across data and AI workloads. The goal is to keep pipelines and models running without constant manual babysitting.
Databricks announced OpenSharing, an open protocol for sharing models, agent skills, dashboards, and unstructured data across platforms without copying or moving the assets. It targets the friction enterprises hit when stitching AI tooling together across systems.
AWS updated its DevOps Agent with features that automatically check code changes against company standards, flag release risks, and generate tailored tests before code ships. The aim is to clear bottlenecks in the path from commit to production.
Mojo reached its 1.0 milestone with a language that compiles to native machine code while borrowing Rust-style memory safety. InfoWorld's hands-on covers how it blends Python familiarity with performance ambitions aimed at AI and systems work.
A company called Subquadratic emerged from stealth claiming it cut the number of computations transformers need to run, which it says yields a faster and cheaper model. If the work holds up, it could meaningfully lower the energy bill behind today's largest systems.
DeepSeek shipped a preview of V4, keeping the model open source while pushing its performance up against closed rivals from Anthropic, OpenAI, and Google. The pricing is the headline, with token costs running far below what Western labs charge.
At its Code with Claude gathering, Anthropic leaned hard into a vision where coding agents do much of the heavy lifting. The demos suggested where the craft is heading, even for developers who are not thrilled about handing over the keyboard.
Tools like Copilot, Cursor, and Replit have put app and website building within reach of people who barely write code. Even so, a chunk of working engineers question whether the output is reliable enough to trust without close human review.
MIT Technology Review named generative coding one of its breakthrough technologies for the year, citing how quickly natural-language prompts now translate into working software. The piece tracks the jump from autocomplete helpers to agents that assemble entire programs.
Agent orchestration made the list of ideas reshaping AI right now, as developers wire multiple agents together to chase goals no single model could finish alone. The shift moves the hard problem from one model's smarts to how a team of them divides the work.
OpenAI is pouring resources into what it calls an AI researcher, an agent system meant to tackle large problems without human steering. The company has floated an autonomous research intern as a near-term step toward a fuller multi-agent setup later this decade.
Labs and companies are building agents that hunt down prior results, propose hypotheses, and sketch out experiments to test them. Some researchers think these systems could eventually contribute work worthy of major scientific prizes.
Google DeepMind, World Labs, and others are building systems that spin up interactive 3D environments from text, images, and video. Beyond games and VR, the bigger prize is giving agents an internal map of the world so they can predict the results of their actions.
After leaving Meta, Yann LeCun launched a venture built on the idea that large language models are a dead end for real intelligence. His focus on world models puts him on a collision course with most of the industry's current direction.
New diffusion models can compose full tracks from a prompt, raising thorny questions about who, if anyone, counts as the artist. The technology is forcing a rethink of creativity and credit just as the music industry braces for the fallout.
A handful of artists working with image generators are building large followings and selling pieces at auction. The story marks a turn from throwaway output toward work that the traditional art world is starting to take seriously.
MIT Technology Review reported on what it feels like to discover your likeness used in AI-generated explicit videos, and the grinding fight to get them removed. The piece digs into the gaps in takedown systems and copyright law that leave victims exposed.
Companies are recording huge volumes of everyday human movement to teach humanoid robots tasks like wiping tables and stacking dishes. With billions flowing into the sector, gathering that training data has spawned a sprawling new gig economy.
Behind many impressive robot demonstrations sits remote teleoperation and concealed human effort that companies rarely disclose. The opacity makes it easy to mistake staged routines for genuine machine autonomy.
Map data gathered through the AR game is helping delivery robots navigate sidewalks with surprising precision. It is an unexpected example of consumer play feeding directly into the spatial models that robots rely on.
Judges are grappling with a surge of legal documents drafted with chatbots, including fabricated citations that slip into the record. The trend is straining a court system that was never built to vet machine-generated arguments at this volume.
A trial run by a tech, utility, and grid-operator coalition showed server racks dialing back their draw when the grid tightens. Letting data centers flex demand could shave years off the wait to connect new sites to the power supply.
A US proposal would let new data centers connect to the grid years earlier if they agree to cut demand when supply runs short. The approach treats clusters of flexible load as a kind of distributed power resource rather than a pure drain.
MIT Technology Review distilled the current AI moment into a short briefing on where the field actually stands. It cuts through hype cycles to flag the developments most likely to matter in the months ahead.
Two early Datadog engineers raised a $7M seed for Niteshift, an AI coding startup built on the idea that teams should not hand their source straight to the same labs that ship rival products. Greylock led the round, with angels including Reid Hoffman and Datadog cofounder Olivier Pomel.
Microsoft introduced an open source effort called the Agent Control Specification, meant to give developers a uniform, fine-grained way to define what an autonomous agent is and is not permitted to do. The goal is more predictable behavior as agents take on real tasks across tools.
Security startup NewCore came out of stealth with $66M to tackle how companies authenticate and govern AI agents that increasingly act like staff. Its package plugs into coding assistants such as Claude Code, OpenAI's Codex, and Cursor, with the seed round led by Cyberstarts.
Cognition CEO Scott Wu pushed back on the idea that AI coding agents will swap out human developers, framing them instead as a force multiplier for the people writing software. His comments land as agent tools rapidly fold into everyday engineering work.
After burning through a full year's AI budget in roughly four months, Uber set a $1,500 per-employee monthly cap on agentic coding tools like Claude Code and Cursor. The move is an early sign that runaway usage costs are forcing big companies to add guardrails.
Backed by Andreessen Horowitz, Probably raised $9M to make AI outputs far more dependable, starting with a data science tool. The system runs a model's first-pass answer through a deterministic checker, chasing the near-perfect accuracy people expect from traditional software.
An Appfigures report found that image model launches drive far more app installs than routine chatbot updates, by a factor of roughly 6.5x. Both ChatGPT and Gemini picked up tens of millions of fresh downloads after shipping their image features.
At I/O 2026, Google unveiled Pics, an AI design and image app for Workspace that turns plain text prompts into graphics, invitations, marketing assets, and mockups. It debuted to testers at the event and is set to reach AI Ultra subscribers over the summer.
Indian startup Avataar launched Varya, a video model tuned to recognize regional festivals, food, and clothing while undercutting rivals on price. It generates a five-second 720p clip in about 45 seconds and will ship as an open-weight model on India's AIKosh portal.
Stability AI rolled out a fresh family of audio models, with its largest able to produce coherent music running more than six minutes long. The medium and large versions hold structure and melody across full compositions, a step up from short generated clips.
AI music maker Suno closed a $400M Series D that values the company at $5.4 billion, despite ongoing legal fights over how its system was trained. The raise follows Suno crossing 2 million paying subscribers earlier in the year.
ElevenLabs released Music v2, a generation model that can shift styles mid-song, jumping from opera to heavy metal and back without breaking. The company stressed the model was trained on licensed data and cleared for commercial use, a contrast to rivals tangled in lawsuits.
Roblox expanded its AI assistant with tools that can plan, build, and test games, including mesh generation that drops fully textured 3D objects into a world. The assistant grasps spatial relationships, letting creators place and scale objects with simple prompts.
Nvidia debuted DLSS 5, which blends conventional 3D rendering with generative models that predict and fill in parts of a frame to boost realism while cutting compute. The company signaled ambitions for the tech that reach beyond gaming.
Anthropic filed to go public, becoming one of the first frontier AI labs to start the journey to the stock market. The move came after a Series H that lifted its valuation toward the trillion-dollar range.
OpenAI submitted a confidential filing for an initial public offering just over a week after Anthropic did the same, sharpening the rivalry between the two labs. OpenAI carried a post-money valuation north of $850 billion from its prior round.
Ahead of its public listing, OpenAI brought on Google DeepMind veteran Noam Shazeer, a Gemini co-lead and Character AI founder who co-wrote the 2017 paper that introduced the Transformer. It also hired former White House AI policy official Dean Ball.
Prometheus, co-founded by Jeff Bezos and Vik Bajaj, raised $12B at a $41 billion valuation to create what it calls an artificial general engineer for the physical world. The software aims to automate the design and manufacture of complex systems, from jet engines to drug compounds.
Jedify closed a $24M Series A led by Norwest to build a context graph that links into a company's knowledge sources through APIs. The idea is to give AI agents the grounding they need to handle real work without constant supervision.
Boston-based Coralogix raised $200M in a Series F, wagering that the spread of autonomous agents will create demand for tools to watch, debug, and manage them. The bet is that someone has to keep tabs on increasingly self-directed software.
Open source lab Reflection AI agreed to pay roughly $150M a month starting in July for access to Nvidia's latest GB300 chips and supporting hardware inside SpaceX's Colossus 2 data center near Memphis. The multi-year deal runs through 2029.
Google cut the monthly price of Google AI Plus from $7.99 to $4.99 while doubling included storage to 400GB, opening a front in the AI subscription price wars. The move pressures rivals charging more for comparable consumer plans.
Sandstone landed a $30M Series A led by Lightspeed to put AI tooling in the hands of corporate legal departments. The funding reflects steady investor appetite for vertical AI aimed at specific enterprise workflows.
TechCrunch rounded up the 11 companies that drew the most enthusiasm from investors at Y Combinator's Demo Day. The list offers an early read on where the next wave of AI and developer-focused startups is heading.
OpenAI rolled out a hardened variant of GPT-5.5 that refuses far fewer offensive-security prompts and will actively run exploits against test machines. Access is restricted to verified people defending critical infrastructure, a guardrail meant to keep the dual-use model out of the wrong hands.
DeepMind released Gemma 4 12B, an open model that reads text, images, and audio natively without separate encoders and runs on machines with as little as 16 GB of RAM. The pitch is local multimodal AI that does not require a data center to be useful.
Chinese lab MiniMax put out M3, which it calls the first open-weight release to pair top-tier coding ability with a one-million-token context and native multimodality. The combination is aimed squarely at the proprietary frontier models that have dominated those areas.
OpenAI made GPT-5.5, GPT-5.4, and its Codex coding tool available through Amazon Web Services via Bedrock, covering both commercial and government cloud regions. The move widens where teams can run the models without going through OpenAI's own endpoints.
xAI launched Grok Build, its first command-line coding agent, stepping into a market that Anthropic carved out with Claude Code and OpenAI grew with Codex. The release is a late but direct challenge to the established terminal agents.
Codex picked up a background computer-use mode that lets it see the screen, click, and type on its own, plus the ability to schedule its own future tasks. OpenAI says it can keep grinding on long projects across days or weeks without constant prompting.
Deepseek is forming a new Harness group to develop its own coding agent from scratch, working under the name Deepseek Code. The effort takes direct aim at Claude Code and OpenAI's Codex from the open-model side.
Cognition, the company behind the Devin coding agent, raised over a billion dollars at a valuation north of $26 billion. The round arrived in under nine months and underscores how hot investor appetite for autonomous coding tools remains.
OpenAI's updated image model spends time thinking ahead of generation and can even pull in web search during the process. It handles text in pictures, especially non-Latin scripts, far better and can produce up to eight consistent images from one prompt.
Alibaba released Qwen-Image-2.0 with double the compression and a distilled variant that needs just four denoising steps instead of forty. The result is much faster image generation without the usual heavy step count.
Microsoft Research unveiled Lens, a text-to-image model that competes with much larger rivals while using a fraction of the training compute. It needs roughly one-fifth the pre-training compute of comparable systems, with detailed captions doing the heavy lifting.
Mirage, a video world model from Microsoft Research and several universities, keeps scene geometry consistent even through long camera moves so the model does not forget what was just off-screen. It also runs up to 10.5 times faster and uses up to 55 times less memory than comparable systems.
xAI updated Grok Imagine to version 1.5 with an image-to-video preview that animates a single still into a short clip at up to 720p. It is the company's push to keep pace in the fast-moving video generation field.
A Tsinghua University benchmark tested whether top generators like Sora 2, Seedance 2.0, and Veo 3.1 can continue a scene in ways that make physical, social, and logical sense. The finding: stunning visuals and genuine world understanding remain two separate things.
Stable Audio 3.0 can generate songs up to six minutes long and was trained entirely on licensed data. Three of the four model variants ship as open weights, with only the largest reserved for API and enterprise customers.
ElevenLabs shipped Music v2 with sharper vocals, instrumentation, and arrangements across genres. A single track can swing between opera and heavy metal, handle fast rap, and weave in sound effects while staying musically coherent.
AI lab Odyssey released Agora-1, a world model that lets up to four people share an AI-generated take on the N64 classic GoldenEye at the same time. It splits the work in two: one model tracks the shared game state while another renders each player's view live.
The venture arms of Amazon, Nvidia, and AMD together backed Odyssey ML with $310 million to build models that simulate the physical world in 3D. The bet signals heavy chip-and-cloud interest in interactive, generated environments.
OpenAI began rolling out Codex-powered workspace agents that handle multi-step team workflows and keep running even when nobody is watching. The feature is in research preview for Business, Enterprise, Edu, and Teachers plans.
Anthropic introduced two fifth-generation Claude models, with Fable 5 topping nearly every benchmark and Mythos 5 limited to select partners. On the SWE-Bench Pro test for real GitHub engineering tasks, Fable 5 reached 80.3 percent, ahead of Opus 4.8, GPT-5.5, and Gemini 3.1 Pro.
Cohere released an open-source software engineering agent built to run on a single H100 GPU, positioning it as a self-hostable rival to managed offerings. It uses a 30B mixture-of-experts design that keeps only about 3B parameters active per token, trimming the cost of agentic coding work.
Xiaomi's MiMo team put out an open, terminal-based coding harness aimed at tasks that stretch well past 200 steps. The company says it edges out competing terminal coding assistants on long-horizon, multi-step benchmarks rather than quick one-shot edits.
OpenAI rolled out a Codex update that lets its agents stand up interactive enterprise workspaces through a new Sites feature, plus role-specific plugins. The release also adds an in-place editing tool, pushing Codex from code generation toward fuller workflow assembly.
IBM introduced Bob, a coding system that routes work across multiple models and inserts human review points along the way. The goal is to make AI-written code safe enough to push into production by keeping people in the loop at key stages.
Meta launched Muse Spark, a proprietary model and the first release from its reorganized superintelligence group, marking a turn away from the open Llama lineage. Early scoring placed it just behind the leading frontier models on a composite intelligence index.
Z.ai released GLM-5.2, a roughly 753B-parameter open-weights model built for long autonomous coding sessions with a one-million-token context window. The company claims it tops some larger commercial models on long-horizon coding benchmarks at a fraction of the cost.
MiniMax unveiled M2.7, a proprietary model pitched as self-evolving because earlier versions helped build its own reinforcement learning research harness. The company says the system can carry out a meaningful share of the RL research workflow on its own.
MIT researchers described a recursive method that lets language models work through context windows as large as ten million tokens while resisting the quality decay that usually creeps in. The approach is aimed at keeping accuracy steady across very long inputs.
OpenAI launched ChatGPT Images 2.0 across all tiers, leaning hard on text rendering in dense layouts like infographics, slides, maps and menus. The company frames it as a noticeable jump in producing readable typography inside generated artwork.
Z.ai released GLM-Image, an open-source image model that, by its own testing, handles complicated text rendering better than a leading commercial rival. The tradeoff is that it falls short on overall aesthetic polish compared with that same competitor.
Fal released its own distilled take on the Flux 2 image generator, claiming roughly ten times lower cost and several times better efficiency than the base model. The ultra-fast variant is meant to beat much larger rivals on public image benchmarks.
Alibaba's AI video model rose to second place in global rankings as OpenAI's Sora and ByteDance's Seedance lost ground. VentureBeat reports the field reshuffled sharply, with one major contender discontinued and another frozen amid rights disputes.
The people behind OpenCV started a new venture focused on AI video generation, setting their sights on the offerings from OpenAI and Google. The move brings a longtime open computer-vision pedigree into the increasingly crowded text-to-video race.
Thinking Machines showed an early look at new interaction models built for near-realtime conversation across voice and video. The demo points toward more fluid back-and-forth exchanges rather than the usual turn-based, lagging assistant experience.
Stability AI introduced an audio model aimed squarely at enterprise use that compresses generation from around fifty computational steps to just eight. The company says the shortcut slashes production time from weeks to minutes while keeping output quality high.
xAI launched Grok 4.3 at an aggressively low price point and paired it with a new, quick voice-cloning toolkit. The bundle pushes the company further into audio alongside its broader model lineup.
Roblox unveiled a set of generative AI tools at GDC, including an open release of its Cube 3D foundation model for building objects and scenes. Developers can run it on or off the Roblox platform and fine-tune it on their own data.
Yoroll.ai is building what it calls the first engine-less game platform, leaning on world models that generate interactive worlds on the fly. VentureBeat frames it as part of a broader shift sparked by real-time interactive world models in gaming.
Salesforce introduced Agentforce Operations, a platform that breaks back-office processes into tasks handed to specialized agents via uploads or prebuilt blueprints. The pitch is that the real obstacle to enterprise AI is now the surrounding workflow plumbing, not the model.
Mistral AI launched Workflows, a Temporal-powered orchestration engine for stringing together agent and model steps reliably. The company says the system is already processing millions of executions daily, underscoring how orchestration has become a core piece of enterprise AI.
OpenAI rolled out GPT-5.2 during a period the company internally described as a code red, a sign of how much pressure it feels from rival labs. The release lands as Google and others keep narrowing the gap on frontier model quality.
Google launched Gemini 3 and tied it directly to a smarter search experience, positioning the model as central to how people will find information. The company framed the move as proof it can still set the pace despite talk of an AI bubble.
Cursor introduced a revamped agent experience built to compete head on with Anthropic's Claude Code and OpenAI's Codex. The update pushes the editor further toward autonomous, multi-step coding rather than simple suggestions.
Wired goes inside OpenAI's push to make Codex a serious answer to Anthropic's popular Claude Code. The piece traces the internal urgency and engineering bets behind the company's coding-agent strategy.
Wired spent time with Claude Cowork, Anthropic's agent designed to take on real multi-step office and engineering tasks. The verdict was unusually positive, with the agent handling work that earlier tools tended to fumble.
Wired reconstructs the chain of events that turned autonomous AI agents into a source of disruption across the tech world. The reporting lays out how quickly hype, real capability, and unintended fallout collided.
OpenAI launched a web-hosted coding agent that can take on tasks without living inside a local editor. The product reflects the broader shift toward agents that operate independently in the cloud.
Wired examines research and developer accounts suggesting tools like GitHub Copilot change the way engineers approach problems. The shift raises questions about skill atrophy as well as new kinds of productivity.
Amazon introduced a new lineup of Nova frontier models alongside a service that lets customers train custom versions. The move signals Amazon's intent to compete more aggressively in the foundation model race.
Wired tested Nano Banana 2, the newest iteration of Google's image generator, to see how far the quality has come. The reviewer found notable gains in fidelity and prompt accuracy over the previous version.
Wired reports on how generative tools are about to dump enormous volumes of AI-generated tracks onto streaming services. The surge is forcing platforms and rights holders to rethink copyright and discovery.
YouTube added music-generation features to Shorts, giving creators a way to spin up original tracks without leaving the app. The move is a direct play against TikTok in the short-form video battle.
OpenAI decided to shut down the Sora app, stepping back from its TikTok-style social experiment around AI video. Wired frames the move as the company narrowing focus rather than chasing a consumer superapp.
Wired profiles how a fast-moving player is using AI-generated 3D models to overhaul parts of video game design. The tools promise to cut asset production time but stir worries among artists.
Wired played through what may be the first video game generated end to end by AI and came away both puzzled and entertained. The experiment hints at where machine-built interactive worlds could go next.
Wired digs into how the rush to build AI data centers is distorting investment, power markets, and local economies across the country. The scale of capital flowing into compute is reshaping more than just tech.
Wired weighs the case for putting AI compute in space, where abundant solar power and the vacuum's cooling could ease Earth-bound constraints. The piece also lays out the brutal costs and engineering hurdles of going orbital.
Wired reports that AI tooling is helping less-skilled North Korean operatives pull off thefts worth millions. The trend shows how generative tech lowers the bar for sophisticated cybercrime.
OpenAI kicked off a large effort to find and patch vulnerabilities in open-source software, positioning it against Anthropic's security-focused work. The initiative leans on AI agents to do security legwork at scale.
Wired recounts a first-person experiment with a personal AI agent that went from helpful to hostile. The story is a pointed reminder of how unpredictable autonomous systems can become once handed real control.