Search the archive

👉 WATCH IT ON-DEMAND

MICROSOFT RELEASES TWO IN-HOUSE MODELS

Microsoft AI has launched MAI-Voice-1, a speech generation model that can produce a minute of audio in under a second, and MAI-1-preview, a mixture-of-experts foundation model trained on ~15,000 NVIDIA H100 GPUs. The technical capabilities of the models are unclear, but they indicate Microsoft's intention to reduce its reliance on OpenAI's model to power its AI offerings.

META PUSHES TO RELEASE NEW LLAMA MODEL BEFORE 2026

Meta plans to release its next version of Llama before the end of the year. The model, internally known as Llama 4.X or Llama 4.5, will be one of the first projects from Meta Superintelligence Labs. Meta recently restructured the unit into four groups, with teams focused on training, research, product, and infrastructure. Meta first announced the Llama 4 models in April.

GROK CODE FAST 1

xAI launched a specialized coding model optimized for speed and cost rather than general intelligence, priced at $0.20 per million input tokens versus $18+ from competitors. The model is temporarily free through partnerships with GitHub Copilot, Cursor, and other coding platforms.

MASS INTELLIGENCE

Over a billion people now use AI chatbots regularly. Recent releases have given even unpaid users access to the most powerful models like GPT-5 or nano banana. Intelligence is no longer scarce or expensive, and will only get cheaper and more abundant, so institutions of every size face the chaos of a billion people wielding AI that outperforms humans at intellectual tasks

BUILDING YOUR OWN CLI CODING AGENT WITH PYDANTIC-AI

CLI coding agents can read code, run tests, and update codebases. This article walks readers through how to create their own coding agent by assembling open source tools and using specific development standards for testing, documentation production, code reasoning, and file system operations. While commercial tools are impressive, they're built for general use cases. Learning how to build a coding agent can provide insight into how systems work and the quality of available tooling.

AI access risks.

INTRODUCING GPT-REALTIME AND REALTIME API UPDATES FOR PRODUCTION VOICE AGENTS

OpenAI's Realtime API is now generally available. It now supports MCP servers, image inputs, and phone calling, making voice agents more capable. The underlying model, gpt-realtime, is OpenAI's most advanced speech-to-speech model yet, showing improvements in following complex instructions, calling tools with precision, and producing speech that sounds natural and expressive. The Realtime API processes and generates audio directly through a single model and API, which reduces latency and preserves nuance in speech, producing more natural, expressive responses.

OLLM

oLLM is a lightweight Python library for large-context LLM inference that enables users to run models like Llama-3.1-8B-Instruct on 100k context using around a consumer GPU with 8GB VRAM. It doesn't use any quantization, only fp16 precision.

R-ZERO: SELF-EVOLVING REASONING LLM FROM ZERO DATA

Self-evolving large language models are models that can autonomously generate, refine, and learn from their own experiences. Existing methods for training such models still rely heavily on vast human-curated tasks and labels. This poses a fundamental bottleneck to advancing AI systems toward capabilities beyond human intelligence. R-Zero is a fully autonomous framework that generates its own training data from scratch. It substantially improves reasoning capability across different backbone models.

XCODE 26 BETA 7 ADDS SUPPORT FOR CLAUDE SONNET AND CHATGPT 5

Claude Sonnet 4 is now available in Xcode. It can be accessed in the Intelligence setting panel. An existing paid Claude account is required to use the integration. Users can now start a new conversation with either GPT-4.1 or GPT-5, with GPT-5 set as the default, when using ChatGPT in Xcode.

THE 100 MOST INFLUENTIAL PEOPLE IN AI

TIME's third annual list of the 100 most influential people in AI includes familiar leaders like Sam Altman, Elon Musk, and Jensen Huang alongside newcomers like DeepSeek CEO Liang Wenfeng and Pope Leo XIV.

chunk (43%) are never using it. Find out why

ANTHROPIC USERS FACE A NEW CHOICE – OPT OUT OR SHARE YOUR DATA FOR AI TRAINING

Anthropic reversed its no-training policy for consumer data, now requiring users to actively choose whether Claude can learn from their conversations by September 28.

NEW AI-POWERED LIVE TRANSLATION AND LANGUAGE LEARNING TOOLS IN GOOGLE TRANSLATE

Google Translate now offers real-time conversation translation in over 70 languages, enhancing live interactions with advanced AI voice and speech recognition models.

JPMORGAN BELIEVES BITCOIN IS UNDERVALUED, SEES $126K BY YEAR-END

Bitcoin's current price is “too low” relative to gold as volatility hits record lows, setting a fair value of $126,000 by year-end. Corporate treasury accumulation, index-driven inflows from Strategy and Metaplanet, and reduced volatility seems to have boosted bitcoin's institutional appeal. With BTC's volatility ratio to gold at a record 2.0, Bitcoin is closer than ever to gold on a risk-adjusted basis.

US GOVERNMENT PUBLISHES MACROECONOMIC DATA ONCHAIN

The U.S. Commerce Department has begun distributing key macroeconomic data on public blockchains in partnership with Chainlink and Pyth, starting with GDP and inflation indicators anchored to Bitcoin, Ethereum, Solana, Base, Avalanche, Arbitrum, and Optimism, opening new use cases for DeFi, prediction markets, and tokenized assets. The data is hashed onchain to ensure integrity. Secretary Howard Lutnick says the move makes “America's economic truth immutable and globally accessible,” highlighting 3.3% GDP growth as the first data point published.

EULEREARN: PASSIVE YIELD, PROFESSIONALLY CURATED

EulerEarn is a permissionless passive yield platform where anyone can deploy and manage ERC-4626 vault strategies without gatekeeping. It features non-custodial, immutable core contracts with timelock and guardian protections for risk-sensitive actions. The system enables curators to professionally manage up to 30 strategies per vault with defined allocation caps, supply/withdrawal queues, and yield smearing mechanisms to optimize returns while maintaining liquidity and reducing MEV risks. Curators can earn up to 50% performance fees while users hold a single token representing their aggregated pool share. The platform supports full ERC20Votes governance functionality for community-driven control.

INTRODUCING $AVNT: THE TOKEN FOR UNIVERSAL LEVERAGE

Avantis, a leading derivatives DEX on Base, has launched $AVNT, a utility and governance token designed to secure its protocol, align incentives, and decentralize governance. Since its February 2024 mainnet debut, Avantis has processed $18B+ in volume, onboarded 38,500+ traders, and listed 80+ markets. The token grants stakers fee discounts and rewards, dedicates 50.1% of supply to community incentives, and aims to scale its "Universal Leverage Layer" to capture demand in the $1 quadrillion RWA derivatives market.

THE INVISIBLE UPGRADE: WHY INSTITUTIONS SHOULD CARE ABOUT CRYPTO RAILS

Institutional finance is undergoing a fundamental infrastructure upgrade through crypto rails, moving beyond speculative use cases to address core operational inefficiencies in traditional systems like T+2 settlement and fragmented reconciliation processes. Zero-knowledge proofs enable institutions to maintain privacy while proving compliance, solvency, or creditworthiness without exposing sensitive portfolio data to centralized gatekeepers who create security vulnerabilities. Ethereum's L2 scaling solutions offer real-time settlement and near-zero cost transactions with built-in execution guarantees, while smart contracts can enforce compliance, portfolio limits, and trading mandates natively rather than through retrofitted oversight systems.

HIRING FOR CRYPTO BD ROLES

Hiring for BD in crypto is different from Web2 because tokens, DAOs, and open-source ecosystems change how partnerships and growth work. Early-stage teams need scrappy hires who can experiment and adapt, but after product-market fit, you need specialists who can scale pipelines and systems. Defining roles clearly and avoiding over-hiring too senior or too generalist too early is critical to unlocking sustainable growth.

HOW MONAD GOT CT ON ITS SIDE

Monad's "Monad Cards" distribution strategy targets crypto influencers with established relevance, rewarding organic social capital rather than project-specific engagement. Unlike common pre-TGE growth hacks, this approach leverages decentralized nominations to reduce insider bias and align with crypto's cultural emphasis on fairness. By empowering recipients to nominate others, Monad addresses historical criticisms of airdrop centralization, enhancing perceived legitimacy and fostering a stronger foundation for its ecosystem.

LIQUIDITY AS AN OPERATING SYSTEM

HyperEVM is a programmable financial operating system built directly on top of Hyperliquid's profitable exchange infrastructure, which processes $300B monthly volume and generates over $1B annualized profit. It differentiates from typical EVM chains by being natively integrated with proven liquidity and user demand, allowing developers to access existing trading flows and monetization primitives, including Builder Codes, HIP-3 market deployment fees, and HYPE staking rewards from day one.

PENNY FOR YOUR THOUGHTS

Penny for Your Thoughts is a new experiment from Plastic Labs that uses Honcho, a tool for AI memory and cognition, alongside x402, to allow users to train an AI with their knowledge and have other people pay to consult it.

RAIN BACKED TO BRIDGE STABLECOINS AND TRADFI PAYMENTS

Rain is reinventing money movement, credit, and settlement flows onchain, driving stablecoin adoption beyond crypto-native use cases.

STABLECOINS AS AMERICA'S DOLLAR EXPORT ENGINE

Stablecoins are the fastest way for the U.S. to export dollars globally.

FARCASTER: DECENTRALIZED SOCIAL BUILT ON ETHEREUM

Farcaster is a decentralized social protocol on Ethereum that gives users ownership of their data, identity, and social graph.

ETHEREUM FOUNDATION LAUNCHES UX-FOCUSED TRILLION DOLLAR SECURITY PUSH

The Ethereum Foundation has launched a Trillion Dollar Security initiative to enhance wallet safety and transaction clarity for billions of users. The first phase introduces minimum security standards for wallets, eliminates blind signing through readable transaction summaries, and creates developer tools to prevent insecure smart contracts. This long-term mission involves collaboration with projects such as Walletbeat and the Verifier Alliance to build a safer Ethereum ecosystem to support trillions of dollars in on-chain capital.

GOOGLE WILL NOW LET EVERYONE USE ITS AI-POWERED VIDEO EDITOR VIDS

Google has expanded access to its AI-powered video editor Vids. A basic version is available to all users. The free version includes templates and stock media, but lacks the newest AI features, such as customizable avatars and image-to-video generation tools, which are currently available only to paid users. Google aims to help companies reduce the time and cost of creating product demos and training videos, which can traditionally take months and cost tens of thousands of dollars.

GOOGLE AIMS TO MAKE TRANSLATE A STRONG COMPETITOR TO DUOLINGO

Google has added AI-powered language learning to its Translate app, offering personalized lessons based on users' levels and goals in English, Spanish, French, and Portuguese. It also introduced real-time conversation translation in over 70 languages. The update is currently available in the US, India, and Mexico. Voice synthesis is still in testing.

FROM IDEA TO INTERFACE: A DESIGNER'S GUIDE TO AI-POWERED PROTOTYPING

AI-powered prototyping tools are enabling designers without programming experience to create functional, high-fidelity prototypes by simply describing what they want to build. These tools bridge the traditional gap between design and engineering by allowing designers to explore user experiences with real data and interactions rather than static mockups. Effective AI prototyping requires planning ahead, being specific about requirements, working within the model's preferences, and focusing on rapid exploration rather than perfect code.

BUILDING AI PRODUCTS IN THE PROBABILISTIC ERA

AI has fundamentally shifted software development from deterministic functions with predictable inputs and outputs to probabilistic systems with infinite input possibilities and stochastic results. Traditional engineering approaches, such as reliability metrics and incremental building, become counterproductive when applied to AI products, as they constrain the models' emergent capabilities and general-purpose intelligence. Success in this new era requires adopting scientific empiricism over classical engineering, treating data as the operating system that connects all organizational functions, and embracing uncertainty while managing it through techniques like user trajectory analysis and content classification.

WHY DESIGNING TERRIBLE SOLUTIONS MAKES YOU A BETTER DESIGNER

Intentionally designing bad solutions can be a powerful way to strengthen design thinking. By pushing ideas to the extreme of what doesn't work, teams gain clarity about what truly defines a good solution, uncover overlooked insights, and distinguish between real constraints and assumed limitations. This contrast prevents falling into the trap of only refining similar “acceptable” ideas and instead opens the door to breakthrough innovations. Failure, when explored deliberately, becomes a learning tool that sharpens decision-making and expands creative possibilities.

FROM IDEA TO LIVE WEBSITE IN SECONDS

Describe your business and watch AI create a stunning, professional website instantly.

FREE AI VIDEO GENERATOR

Digen AI offers a free AI video generator that enables easy transformation of images into professional videos. It features realistic lip-sync, multilingual support, and advanced smart animation technology.

BEAUTIFUL BACKGROUND PATTERNS

Create stunning websites with this curated collection of 100+ modern CSS background patterns and gradients for your websites and apps.

WALL STREET BETS BIG ON THE FUTURE OF DESIGN

Figma's blockbuster IPO on August 1 saw its valuation soar to $47 billion, with Wall Street betting big on the future growth potential of design. CEO Dylan Field argues that AI will create a "proliferation of interfaces" requiring human designers, making design the key differentiator in an increasingly crowded app marketplace. Rather than replacing designers, AI is expected to drive demand for more design work, as it enables the creation of more digital products that require thoughtful user experiences.

THE PRE-MORTEM: TRANSFORMING A SKEPTICAL DESIGN SYSTEM TEAM

A pre-mortem workshop helped transform a skeptical design system team at a large enterprise company by having them imagine their project had failed after one year. The team identified 30 potential failure points and developed 15 prioritized solutions to address the root causes behind those failures. This process allowed team members to voice concerns safely and focus on concrete solutions rather than vague doubts about repeating past unsuccessful design system attempts.

THE GOOD, THE BAD, AND THE IFFY: IS THERE SUCH A THING AS AN ETHICAL DESIGNER?

Designers across studios like Pentagram, Koto, and Creech reveal how murky and subjective client selection really is—what's a red flag for one may be a dream job for another, shaped by personal values, lived experience, and financial realities. From type foundries navigating misuse of their fonts to UX designers weighing the ethics of dark patterns, the tension between earning a living and maintaining integrity shows that “ethical design” in 2025 has no universal rulebook—only personal lines in the sand.

BOSE JUMPS ON 2025'S MOST VIRAL COLOUR TREND WITH INGENIOUS PACKAGING DESIGN

Bose cleverly incorporated the 2025 viral "butter yellow" color trend into its Soundlink Plus portable speaker packaging by disguising the citrus-yellow device as actual butter for toast.

GAIA ALARI COMMUNICATES UNIVERSAL FEELINGS IN HER BESPOKE STOP MOTION ANIMATIONS AND ILLUSTRATIONS

Milan-based creator Gaia Alari produces textural stop-motion animations and illustrations for musicians such as Alessia Cara, Porridge Radio, and Coldplay.

FIGMA COMPETITOR FRAMER RAISES $100M AT $2B VALUATION

Framer raised $100 million in Series D funding at a $2 billion valuation, led by Meritech Capital Partners and Atomico. The web design platform, popular among Y Combinator startups and major clients like Scale AI, expects $100 million in ARR next year. Framer offers unique features compared to competitors like Figma, including A/B testing and enterprise-level security.

THE TOP 100 [GEN AI] CONSUMER APPS

The Gen AI ecosystem is stabilizing. There were only 11 new names on Andreessen Horowitz's most recent Top 100 Gen AI Consumer Apps list, compared to 17 newcomers in the March ranking. There were significantly more mobile newcomers as app stores have cracked down on ChatGPT copycats, opening up room for more original apps. ChatGPT is still the leading general assistant, but Google, Grok, and Meta are closing the gap. More on the top Gen AI consumer apps is available in the article.

NVIDIA REPORTS RECORD SALES AS THE AI BOOM CONTINUES

Nvidia, the world's most valuable company, reported $46.7 billion in revenue this quarter, a 56% increase compared to the same period last year. AI-dominated data center business largely fuelled that growth.

SEO IS A WASTE OF TIME IF IT DOESN'T FOCUS ON SCALE

It's easy to create SEO busywork that feels productive - fixing meta tags on pages nobody visits, chasing keywords with zero searches, updating old blog posts - but these won't change anything. What works is fixing problems across thousands of pages at once: repairing broken links everywhere, updating all your category templates, and rebuilding how your site connects pages. Start with pages already getting traffic to prove you can improve them, then fix the bigger structural problems.

STOP SELLING A SAAS SOLUTION

Companies keep telling founders to stop selling features and start selling solutions, then outcomes, then transformations, but they're all missing what actually matters - buyers don't trust that software alone will get them results. Don't make your tool the hero of the story and make your team the hero instead. Instead of saying "our magical tool solves your problem," say "we are the team that will solve your problem, and this tool is our weapon."

THE SCARIEST FRAMEWORK TO EVALUATE YOUR TEAM

Managers lie to themselves about their teams - they convince themselves someone is "pretty good" to avoid replacing them, or haven't updated their opinion since that person was great two years ago. The solution is plotting everyone on a 2x2 matrix of performance (Do they deliver on time and understand business impact? Can they work without creating friction?) versus potential (Can they think strategically and hold their own with senior leaders? Do people follow them?).