PERSONAL TASTE IS THE MOAT
Taste is the moat when correctness is cheap.
Full-text search across article titles and summaries. Refine with sector, year, content type, and reading-time filters.
Taste is the moat when correctness is cheap.
Sponsored
AI agents are reshaping digital experiences, but securing them requires rethinking identity and access controls built for a human-first world. That's why we built Auth0 for AI Agents [4], now GA.
At least three of the most valuable and highest-profile tech companies are currently preparing to list their shares on the public market. Anthropic, OpenAI, and SpaceX have all taken early steps to go public. Any of these companies would be among the most valuable ever to go public. The listings could create an enormous bonanza for Wall Street, which has seen a slump in IPOs since 2021.
Tesla will stop selling its Full Self-Driving (Supervised) software for a flat rate and instead only offer it as a monthly subscription after February 14. The feature will start at $99 per month. It was previously priced at a one-time payment of $8,000. The technology still requires a human driver at the wheel ready to steer or brake at any time.
SpaceX has announced plans for a potential uncrewed mission to Mars as early as late 2026. The proposed schedule coincides with a rare planetary alignment between Earth and Mars that offers a narrow but valuable launch opportunity. Several core technologies essential to deep space travel remain in development. Starship Version 3 is expected to debut this year. It will feature upgrades that may prove pivotal to long-duration missions beyond Earth orbit.
OTOF-related hearing loss results from a genetic change in a single critical protein. The auditory nerve fails to receive the proper signals from the hair cells in the ear due to the lack of the proper protein, resulting in profound deafness. DB-OTO is a virus-based gene therapy that can deliver a working version of the OTOF gene. It uses a non-pathogenic virus to carry the proper genetic code that produces the protein. In an early trial of the DB-OTO treatment, eleven out of twelve patients developed improved hearing within weeks, and nine patients no longer required cochlear implants.
Sponsored
55% of knowledge workers are putting in more hours but delivering less. The culprit? Well over half say that fragmentation is a productivity killer. Miro's 2025 Momentum at Work Report [9] lays out the costs of tool sprawl and the solutions that get teams back on
Model choice matters for extremely long-running tasks. GPT-5.2 models perform better at extended autonomous work compared to Opus 4.5 and GPT-5.1-codex. Many improvements come from removing complexity rather than adding it. The best system is often simpler than you'd expect. The harness and models matter, but the prompts matter more.
The most compelling reason to hire juniors is AI itself. The biggest barrier to AI transformation has been workforce buy-in. Gen Z leads in AI adoption, and they act as an accelerant, helping colleagues learn AI tools and workflows. AI has lowered the cost of onboarding. If the infrastructure already exists, hiring juniors may be one of the highest-leverage AI bets available today.
Long-horizon agents allow models to take actions and iterate over time. The rate of progress in this area of research is exponential. It is estimated that agents will be able to work reliably to complete tasks comparably to human experts by 2028. People will soon be able to hire agents - one of the litmus tests for AGI.
Humans will lose control over the future if AI capabilities continue to advance. However, if we manage to avoid that, there's a good chance humans will retain a lot of employment during the Cyborg Era, which might last around 10 to 20 years. It is implausible that AI capabilities and available compute continue to increase indefinitely. This is a transitional state.
Anthropic blocking third parties from using the Claude Code API will not convert people back to Claude Code, it will just convert people to other model providers.
This post presents a list of terms used in Gas Town, an AI agent orchestration tool, decoding the definitions used in the introductory paper for developers.
Google is testing a new AI tool with users that connects information from various apps to provide personalized answers in its chatbot.
This post discusses how to run multiple Claude Code sessions in parallel, each in its own fully isolated environment.
AI enables people to just describe what they want, and then apps write the code needed to do it on demand.
Digg has relaunched with a focus on trust signals, transparency in moderation, and defenses against AI-driven spam.
Sponsored
to get you started...
OpenAI signed a multi-year deal with Cerebras worth over $10 billion to receive 750 megawatts of compute through 2028. The partnership aims to boost inference speed for OpenAI's products.
Gemini's Personal Intelligence beta, launching in the US, connects Google apps to personalize user experiences by securely linking Gmail, Photos, YouTube, and Search for tailored assistance. Users control app connections, enhancing the AI's ability to provide specific, informed responses. Privacy is prioritized, ensuring data remains secure and under user control while continuously refining functionality.
Airbnb appointed Ahmad Al-Dahle, former head of GenAI at Meta, as its new technology chief to enhance AI in travel and e-commerce. The company is shifting beyond short-term rentals, integrating new services, AI chatbots, and aiming to become a travel concierge. Al-Dahle's extensive experience with AI at Meta and Apple aligns with Airbnb's strategic focus on tech-driven growth.
GPU workload tuning and hardware-level optimizations reduced latency and cost across Character.ai's systems and doubled production inference speed.
AI coding agents are making software development nearly free, allowing for personalized applications tailored to individual needs. These agents, like Anthropic's Claude Code, can automate small projects that save time. While the technology is still developing, it promises rapid changes in software customization and productivity by 2026.
Sponsored
Shopify and Google co-developed an open standard for AI agents: the Universal Commerce Protocol (UCP) [12]. UCP defines the discovery and negotiation mechanisms between agent and merchant, and makes it simple for humans and agents to work together. Read more on the Shopify
GPT-5.2-Codex is an upgraded version of GPT-5.2 optimized for agentic coding tasks in Codex and similar environments. It supports four levels of reasoning effort settings and has a 400,000-token context window. The model costs $1.75 per million input tokens and $14.00 per million output tokens. It is now available in the Responses API.
MCP Tool Search is now rolling out in Claude Code. The feature allows Claude Code to dynamically load tools into context. Claude Code detects when MCP tool descriptions would use more than 10% of the context, and when triggered, tools are loaded via search instead of preloaded. MCP tools will otherwise work exactly as before.
Ministral 3 is a new family of dense language models (3B, 8B, and 14B) optimized for low-resource environments, offering base, instruction-tuned, and reasoning variants. The models support image understanding and were trained using Cascade Distillation, an iterative distillation and pruning method.
OpenAI is quietly rolling out a feature called ChatGPT Translate that looks very similar to Google Translate. It is unclear which model OpenAI is using for the feature. ChatGPT Translate is free to use and doesn't require a paid account. Users can keep the conversation going after the first translation. A screenshot of the feature is available in the article.
Anthropic blocked opencode from the Claude Code API, causing frustration among users and encouraging them to switch models. This decision risks driving users to other providers by showcasing hostility against users who don't comply with strict usage policies. Blocking opencode is seen as a first step in enforcing rigid control, potentially alienating its user base.
The emerging trillion-dollar opportunity lies in developing "vibes graphs" capturing ambient organizational sentiment. Current enterprise software tracks structured data but misses the nuanced vibes that influence decisions, leading to potential monetary losses. Startups building vibes-native infrastructure are poised to create competitive advantages over incumbents by capturing real-time emotional cues, unlocking deep insights into organizational dynamics.
Sponsored
71% of data and AI teams spend 25%+ of their time on data connectivity. Learn how leaders are spending less time plumbing and more time developing in CData's 2026 State of AI Data Connectivity Report [19].
Google revamped its Trends Explore page with Gemini AI, adding features that surface related search terms, auto-generate comparisons, and suggest follow-up queries.
Steve Yegge's tool, "Gas Town," introduces a complex AI agent orchestration system with unique terminology.
Microsoft has been using Anthropic's AI to power more and more of its own products.
Sponsored
Stablecoins power trillions of dollars in global money-movement, however, the economics are concentrated: a few players keep most yield, set fees, and control user experience—limiting choice and revenue for builders.
Crypto custody provider BitGo filed for a US IPO seeking to raise up to $201 million at a $1.96 billion valuation, offering 11.8 million shares priced at $15-17 on the NYSE under ticker “BTGO” with Goldman Sachs and Citigroup as lead underwriters. The firm holds approximately $104 billion in assets under custody across 1,550+ digital assets for institutional clients. The IPO comes shortly after BitGo received conditional OCC approval in December to operate as a federally chartered national trust bank, representing one of the first major public offerings from a pure-play crypto custody provider.
Coinbase said it cannot support the Senate market-structure bill after the release of a more than 270-page text that raised unresolved issues around tokenized equities, DeFi treatment, provisions that would “kill rewards on stablecoins,” and the role of the SEC. Timing pressures, 70+ filed amendments, new illicit-finance language, and uncertainty that concerns would be addressed led Coinbase to withdraw backing, and the Senate Banking Committee postponed the scheduled markup while negotiators work to fix the problems.
BVNK will provide stablecoin infrastructure for Visa Direct pilot programs, enabling select business customers to pre-fund payouts in stablecoins and to pay recipients in stablecoins (with validators receiving their preferred stablecoin via a backend flow). This demonstrates a deeper Visa and BVNK integration after Visa Ventures' 2025 investment and signals a push to embed tokenized dollars into high-volume money-movement rails for faster, 24/7 global payments.
Noah and NALA partnered to create a cross-border settlement network that converts global USD collections into stablecoins and routes real-time local-currency payouts via Rafiki, enabling instant USD settlement, 24/7 treasury, and rapid local distribution for businesses across Africa and Asia. This partnership targets payroll, merchant payouts, and treasury use cases while bypassing slow, fee-heavy legacy rails.
Vitalik Buterin has published a framework that outlines seven protocol requirements for Ethereum to achieve “ossification” - the ability to function independently without ongoing developer intervention. The first priority is implementing century-scale quantum resistance to prevent future quantum computing advances from compromising user funds and network consensus. Additional requirements include scaling to thousands of TPS through ZK-EVM validation and PeerDAS, implementing statelessness and state expiry, moving beyond ECDSA to fully abstracted accounts, and establishing censorship-resistant block-building mechanisms. Buterin proposed a staged rollout that completes at least one requirement per year.
Crypto's biggest barrier to mainstream adoption is that everything onchain is public by default - wallet balances, transaction history, DeFi positions, and payroll are all visible to anyone. While early crypto users valued this transparency, it's wildly unsafe for normal people, businesses, banks, and governments. Zama is building what it calls “HTTPS for blockchains” using Fully Homomorphic Encryption (FHE) to enable confidential smart contracts, aiming to upgrade crypto from “everything is public” to “encrypted by default” - potentially the most important infrastructure shift since smart contracts.
Excessive pivoting to trading platforms creates overcrowded markets and sacrifices long-term defensibility. Founders should prioritize genuine product development over speculation-driven token dynamics. Imminent crypto market structure regulation will eliminate industry distortions caused by legal uncertainty - such as avoiding transparency and designing economically valueless tokens - by replacing “enforcement roulette” with clear standards.