A report from Analytics India Magazine claims OpenAI spends around $700,000 every day to keep ChatGPT running, raising alarms about potential bankruptcy by late 2024. Despite heavy backing from Microsoft, shrinking web traffic and rising pressure from open-source alternatives like Llama 2 are straining its path to profitability.
NVIDIA just introduced FlexiCubes, a new approach designed to extract high-quality 3D meshes in generative AI and reconstruction pipelines. Acting as a drop-in replacement for traditional marching cubes, it introduces flexible parameters during optimization to capture finer details and produce meshes ready for physics simulations out of the box.
La-Z-Boy just unveiled The Decliner, an AI-powered recliner prototype that drafts creative SMS excuses to ditch last-minute invites. Created as part of the brand's "Long Live The Lazy" campaign, the chair is built for people who want to kick back without feeling guilty. Only three units exist, given away via an excuse-writing contest.
Amazon is testing a generative AI tool with a select group of merchants to automate product listings. Sellers simply feed in keywords, and the AI drafts titles, bullet points, and descriptions. While rivals like Shopify already offer similar features, Amazon is banking on its vast consumer data to craft higher-converting copy, though merchants still need to verify compliance.
Major news organizations just published an open letter urging global lawmakers to enforce transparency around AI training datasets. Signatories like The Associated Press and Getty Images argue that generative models exploit copyrighted journalism without consent or compensation, demanding mandatory rights-holder consent and commercial licensing protections.
OpenAI just rolled out its custom instructions feature to all users, including the free tier on web, iOS, and Android. It lets you set response preferences, tone, and role context once so you don't repeat yourself in every prompt. Steerability is officially becoming a default expectation rather than a paid perk.
The Biden administration just signed an executive order restricting U.S. venture capital and private equity from investing in Chinese AI, semiconductor, and quantum tech. The restrictions specifically target key sectors like chip design software and fabrication tools. The new rules are set to take effect next year following a public comment period.
ElevenLabs just partnered with Inworld to bring its advanced voice synthesis directly into Inworld's AI NPC platform. Game developers can now generate lifelike voices that dynamically adapt to player interactions in real time. It is a big upgrade for immersion, making in-game characters sound far less robotic.
The New York Times just updated its terms of service to strictly prohibit scraping its content to train machine learning or AI models. The ban applies to text, images, and metadata, warning that unauthorized crawlers could face penalties. Basically, publishers are drawing a hard line against tech giants training models on their work for free.
The U.S. Department of Defense just established Task Force Lima, a dedicated generative AI initiative led by its Chief Digital and Artificial Intelligence Office. The task force will assess and integrate large language models across defense operations while actively guarding against training data vulnerabilities and adversarial threats.
Pak 'n' Save's Savey Meal-bot just told shoppers to brew deadly chlorine gas after users tested it with non-food household items. Dubbed an "aromatic water mix," the bot recommended chilling and drinking the toxic concoction, alongside bleach cocktails and poison sandwiches. The supermarket chain says it is now fine-tuning safety controls.
Stability AI just debuted Japanese StableLM Alpha, a 7B-parameter model marking its entrance into the Japanese language space. Trained on 750 billion tokens, the base model is available for commercial use under Apache 2.0, while the instruction-tuned version tops public benchmarks for research.
Anthropic just launched Claude Instant 1.2, filtering the capabilities of its flagship Claude 2 down into its faster, cheaper text model. It gains a 100K token context window, handles formatting and safety better with reduced hallucinations, and bumps coding benchmark scores from 52.8% to 58.7% via the API.
Google is reportedly negotiating with Universal Music Group and Warner Music to legally license artists' voices and melodies for a new AI song generator. The tool would let users make AI tracks while paying copyright holders, giving artists an opt-in choice instead of fighting unauthorized deepfakes.
The White House just launched a two-year AI Cyber Challenge to patch vulnerabilities across America's critical infrastructure. Backed by DARPA alongside OpenAI, Google, Anthropic, and Microsoft, the competition offers nearly $20 million in prizes for autonomous defense systems. Why it matters: It turns foundation models into frontline shields against real-world cyber threats.
Shutterstock is tapping NVIDIA Picasso to let creators generate and light 3D scene backgrounds using text or image prompts. The models, trained on fully licensed data, spit out 8K 360-degree environment maps on demand. Why it matters: It offloads the tedious chore of background mapping and scene lighting, letting 3D artists focus their energy on core hero assets.
NVIDIA just unveiled AI Workbench, a unified developer toolkit that lets teams test and customize pretrained generative AI models locally on RTX PCs and workstations before scaling to the cloud. Alongside the tool, NVIDIA introduced AI Enterprise 4.0 with integrated NeMo support to bridge the gap between desktop experimentation and production-grade deployments.
Researchers just built an AI system that can identify which keys you press on a laptop simply by eavesdropping on typing sounds. The model hit 95% accuracy using a nearby phone recording and 93% over a Zoom call. Why it matters: as microphones become ubiquitous, typing sensitive passwords during video meetings is turning into a genuine cybersecurity hazard.
Stability AI just launched Stable Code Alpha 3B, its first generative AI model tailored for coding. Trained on 560B tokens of code, the suite includes base, instruction, and long-context variants with a 16,000-token window capable of reviewing multiple files at once. Why it matters: It gives developers a lightweight, accessible open-source assistant for everyday programming.
D-ID and ElevenLabs just teamed up to integrate natural-sounding synthetic speech into D-ID's Creative Reality studio. Pro subscribers can now combine avatars and facial expressions with nine premium voices in a single click, making AI video narrators sound far less robotic.
NVIDIA just launched a suite of cloud APIs and developer services for OpenUSD, aiming to make building 3D virtual worlds as universal as the 2D web. The release includes ChatUSD, an LLM copilot that generates USD Python scripts, and DeepSearch for natural language asset hunting. Jensen Huang is placing a massive bet that OpenUSD will anchor the next generation of industrial digital twins.
NVIDIA just partnered with Hugging Face to bring its DGX Cloud supercomputing straight into the world's largest open-source AI community. Developers will soon get one-click access to clusters running eight H100 or A100 GPUs per node through a new Training Cluster as a Service. The integration aims to remove enterprise compute bottlenecks and launches in the coming months.
OpenAI just launched GPTBot, a web crawler designed to collect publicly available web data to train future models like GPT-5. The crawler automatically filters out paywalls, personal data, and policy-violating content. Website owners can opt out by adding a "disallow" command to a standard file on their server.
Microsoft just launched its Azure ND H100 v5 virtual machines, packing eight NVIDIA H100 GPUs and Quantum-2 InfiniBand networking for massive AI workloads. The cloud giant also expanded Azure OpenAI Service to five new regions, broadening access to models like GPT-4 as enterprise adoption surges. Why it matters: Microsoft is rapidly scaling up hardware muscle to meet insatiable cloud AI demand.
Zoom quietly updated its terms of service to train AI models on "service-generated data" like telemetry and product usage, with no opt-out available. While user audio, video, and chats are excluded by default, enabling Zoom's new generative AI meeting tools requires signing consent to share that content for model training.
Google just added an AI-powered grammar checker directly into Search to rival Grammarly. Typing a sentence alongside "grammar check" analyzes your phrasing, highlights corrections, or gives you a green checkmark if it's spot on. It currently only supports English, but it saves you a trip to Google Docs just to proofread.
Microsoft just celebrated six months of its AI-powered Bing, revealing users have logged over 1 billion chats and created 750 million images. The biggest news: Bing Chat is finally expanding to third-party browsers on web and mobile. Why it matters: Dropping the Edge requirement makes Microsoft's flagship AI features accessible to millions who refused to switch default browsers.
A Cointelegraph analysis warns that the generative AI hype is showing classic bubble symptoms after raising over $25B in the first half of 2023. Startups are burning funding just to stockpile scarce Nvidia GPUs rather than building product moats. With ChatGPT traffic slipping and API wrappers laying off staff, the market may face a reality check.
Apple CEO Tim Cook just pushed back on concerns that the company is missing the AI boom, insisting that AI and machine learning are already "virtually embedded in every product" Apple builds. Cook revealed Apple has been researching generative AI for years, pointing to rising R&D spending as direct proof of its massive investment. Why it matters: While rivals rush out flashy chatbots, Apple is signaling that its AI play is quietly woven straight into the devices people already use every day.
NVIDIA CEO Jensen Huang is returning to SIGGRAPH to deliver a live keynote on Aug. 8. He will showcase NVIDIA's latest generative AI breakthroughs and dive into OpenUSD developments following a new alliance with Apple, Pixar, and others. Watch out for what comes next in 3D graphics and visual computing.
Amazon CEO Andy Jassy just told investors that "every single one" of the company's businesses has multiple generative AI projects underway. From AWS and retail to ads and Alexa, Jassy confirmed the technology is now central to their strategy. Why it matters: Amazon is making sure every division catches up to the AI boom, with big upgrades likely coming to Alexa and search soon.
Elon Musk just seemingly snatched the ultra-valuable AI.com domain, which now redirects to his research outfit X.ai instead of OpenAI's ChatGPT. Premium two-letter domains typically command millions of dollars, turning what looks like a simple redirect into a costly flex between AI rivals.
The U.S. Air Force just demonstrated an AI agent flying the stealthy XQ-58A Valkyrie drone through a three-hour combat challenge. Trained on millions of simulation hours, the software successfully handled real airborne tactical tasks. While the Pentagon plans to pair thousands of autonomous drones with crewed fighters, officials stress that humans will remain in the loop.
OpenAI just announced a bundle of practical updates coming to ChatGPT over the next week. Plus subscribers will finally keep their selected model by default instead of reverting to GPT-3.5, and Code Interpreter now supports analyzing multiple files at once. Users also get helpful prompt starters, suggested follow-up replies, longer login sessions, and handy keyboard shortcuts.
Google just partnered with Grammy-winning lyricist Lupe Fiasco to launch TextFX, an AI tool built specifically for writers and rappers. Instead of writing verses for artists, it uses language models to generate alternate meanings, wordplay, and conceptual tangents from chosen words. The goal is to act as a brainstorming assistant rather than replace the artist.
Alibaba just open-sourced two 7-billion-parameter AI models, Qwen-7B and conversational variant Qwen-7B-Chat, supporting both English and Chinese. Global developers can build on them for free, while companies with over 100 million active users will need a royalty-free license. Why it matters: Alibaba is copying Meta's open-source playbook to win over developers and drive adoption of its cloud computing services.
A deep-dive analysis argues that generative AI and foundation models are breaking the historical mediocrity spiral of legacy software. Unlike older systems crippled by long-tail accuracy costs and human-in-the-loop overhead, consumer hits like ChatGPT and Character.AI are unlocking unprecedented organic adoption. Why it matters: Startups finally have the growth velocity and defensibility to challenge entrenched incumbents rather than ending up as bespoke IT services.
IBM and Hugging Face just open-sourced a geospatial AI foundation model built in partnership with NASA. Trained on a year of satellite data across the continental U.S., the model beats state-of-the-art benchmarks by 15% using half the labeled data. Why it matters: It turns massive satellite archives into ready-to-use tools for tracking floods, fires, and deforestation.
Google just rolled out three handy upgrades to its Search Generative Experience (SGE). AI overviews now generate twice as fast as before, search summaries will soon embed how-to videos and images, and reference links now display publish dates so you know how fresh the information really is.
Tinder is testing an AI feature that scans your photo album and automatically picks the five best shots to represent you on your dating profile. Match Group CEO Bernard Kim explained the tool aims to take the stress and awkwardness out of profile creation. The feature is currently in testing alongside plans for AI-assisted bios.
Meta just open-sourced AudioCraft, a unified codebase designed to generate high-fidelity audio and music from simple text prompts. The release packages together MusicGen, the sound-effect generator AudioGen, and an improved EnCodec neural decoder, complete with open weights and code for researchers to train their own models.
Google just pulled its AI Test Kitchen app from both the App Store and Google Play, shifting the experimental playground entirely to the web. The platform, which once showcased early LaMDA demos, now only hosts the MusicLM text-to-audio tool. Google confirmed it is prioritizing the web experience to push updates faster on a single platform.
Instagram is working on notices that flag when content was "created or edited with AI," according to app screenshots uncovered by a researcher. The move follows Meta's recent White House pledge to develop watermarking systems, suggesting the platform may proactively label synthetic images instead of relying purely on user disclosures.
Nvidia researchers just unveiled Perfusion, a lightweight personalization method for text-to-image models that takes up only 100KB. Developed with Tel-Aviv University, it trains in four minutes and uses "Key-Locking" to keep an object's visual identity while avoiding overfitting. Why it matters: It delivers results comparable to multi-gigabyte models like DreamBooth while taking up virtually no disk space.
Kickstarter just announced that starting August 29, any crowdfunding campaign using generative AI must explicitly disclose which parts are AI-generated and which are original. Creators developing AI tools must also share their training data sources and creator consent safeguards, or risk having their projects suspended.
YouTube is testing AI-generated video summaries on search and watch pages to help viewers quickly decide if a clip is worth watching. The test is currently limited to select mobile users on English-language how-to, shopping, and vlog videos, and YouTube notes it won't replace creators' own descriptions.
US lawmakers are finally drafting formal playbooks to police advanced AI, spanning mandatory audits, FDA-style licensing, and chip-level compute tracking. The push targets everything from copyright and bias to rogue models and bioweapons. Why it matters: The wild west of runaway frontier model deployment is about to face Washington red tape.
Meta is preparing to launch AI chatbots with distinct personalities across Instagram and Facebook as soon as next month. Powered by its LLaMA models, the characters reportedly range from a Lincoln impersonator to a chill surfer offering travel advice. Why it matters: The interactive characters aim to fend off TikTok, prove Meta's generative AI chops, and capture rich user interest data for ad targeting.
OpenAI just expanded ChatGPT for Android to all countries and territories where ChatGPT is officially available. The mobile app now matches web availability across dozens of regions. Users accessing the service outside supported areas still risk account suspension.
News Corp Australia is using generative AI to produce 3,000 local news stories every week with just a four-person team. The automated copy covers routine service info like fuel prices and weather across 75 hyper-local outlets, driving 55% of the company's new subscriptions. While journalists oversee the output, readers get zero disclosure that AI helped write it.
OpenAI just filed a trademark application for "GPT-5" with the USPTO, covering everything from natural language processing and speech recognition to artificial neural networks. While the filing signals the company's long-term roadmap, don't expect a release anytime soon—CEO Sam Altman recently noted that the team is nowhere close and still has extensive safety audits ahead.
Google is overhauling its Assistant to integrate generative AI similar to Bard and ChatGPT, according to an internal email sent Monday. The revamp kicks off on mobile devices alongside an internal reorganization and dozens of layoffs. Why it matters: The classic voice assistant era is ending as Big Tech races to replace scripted bots with actual LLMs.
The Open Source Initiative (OSI) just called out Meta for misusing the term "open source" with LLaMA 2. Because Meta restricts certain commercial uses and application areas, the license contradicts the Open Source Definition. OSI argues real open source gives developers full sovereignty, and plans to release a dedicated "Open Source AI" standard draft on October 17.
AI creators just revived digital artist Greg Rutkowski's style using a custom LoRA model on Civitai, bypassing Stability AI's official removal. Rutkowski had campaigned to purge his work from Stable Diffusion 2.0, but open-source creators stepped in to recreate his aesthetic anyway. Why it matters: It shows that in an open-source world, opt-out requests are practically impossible to enforce.
Google DeepMind just unveiled RT-2, a vision-language-action model that translates web data directly into robotic movements. Rather than relying on separate systems for reasoning and motor control, a single model now directs real-world tasks. In tests on novel, unseen scenarios, its success rate jumped from RT-1's 32% to 62%. Why it matters: General-purpose robots just took a major leap forward.
