05-15-2026

The Frontier

Your signal. Your price.

  • 1d ago

    Tan credits the November 2025 release of Anthropic's Opus model as a watershed moment, enabling 'vibe coding' that lets him produce 100x more software now than in 2013. He sees this as democratizing creation.

  • 1d ago

    Tan views Silicon Valley's essence as earnest builders making things people want. He attributes its origin to post-WWII R&D and defense funding, like DARPA's role in creating TCP/IP, which was later commercialized.

  • 1d ago

    Kyle Olney argues the Blockchain Regulatory Certainty Act (BRCA), Section 604, is the existential provision of crypto market structure legislation, as it protects non-custodial software developers from being prosecuted as money transmitters.

  • 1d ago

    Olney warns that without BRCA protections, developers like those behind Tornado Cash and Samourai Wallet face criminal prosecution for publishing code, which would drive innovation offshore to jurisdictions like Singapore or the UAE.

  • 1d ago

    Gustavo notes Bitcoin Core PR #33796 adds BTCK_check_transaction endpoint. It runs context-free consensus checks on transaction structure.

  • 1d ago

    Gustavo explains PSBT v2 (BIP370) is now default in Bitcoin Core. It allows modular transaction construction, supporting adding inputs/outputs mid-process.

  • 1d ago

    BIP451 defines a DUST UTXO disposal protocol. It uses SIGHASH_ALL|ANYONECANPAY, letting anyone batch dust inputs into a single zero-value OP_RETURN output.

  • 1d ago

    Murch notes the protocol's OP_RETURN includes 'ASH'. A legacy input creates a ~65 byte transaction, while a SegWit input creates a transaction of exactly that size.

  • 1d ago

    RustBitcoin adds a V1_message_header constructor. It lets developers build P2P message headers without requiring network transmission.

  • 1d ago

    The new 'extension bolt' Bolt 995 defines simple taproot channels. It uses MuSig2 and excludes gossip announcement specs, which will come in a follow-up.

  • 1d ago

    Zero-fee commitment channels (Bolts 1228) use a 240-sat ephemeral anchor output. The specification caps HTLCs at 114 due to the 10KB transaction size limit.

  • 1d ago

    Bolts 1327 updates RBF logic. It ensures fee bumps meet both Bolt's 25/24 multiplier and an absolute 25 sat/kwu minimum, aligning with BIP125 replacement rules.

  • 1d ago

    Zach Herbert says Foundation's AI integration has accelerated their development pace significantly, though AI models still struggle with low-level firmware and driver code.

  • 1d ago

    Herbert blames legacy operating systems like Mac OS, Windows, and Linux, built on 30-year-old Unix code with massive attack surfaces, for being unable to distinguish between human and AI agent actions.

  • 1d ago

    Foundation's Passport Prime runs on a custom microkernel operating system called KOS, with a kernel under 9,000 lines of code written in Rust, designed for minimal attack surface and app sandboxing.

  • 1d ago

    KOS sandboxes third-party apps via a message-passing microkernel, memory isolation using an MMU, and grants apps only hardened derived child keys - never the master seed.

  • 1d ago

    Herbert warns that AI models like Claude's Mythos will likely expose zero-day vulnerabilities in massive codebases like the Linux kernel or Chromium weekly, making current operating systems untenable for security.

  • 2d ago

    Claude Code's annualized revenue grew from $1 billion to $2.5 billion in about two months during Q1 2026, while its use expanded beyond coding to preview broader agentic trends.

  • 2d ago

    Cursor doubled its annualized revenue to $2 billion this quarter, Lovable hit $400 million ARR with a $100M monthly jump, and Replit projects $1 billion ARR by end of 2026.

  • 2d ago

    AI usage surveys show 71% of practitioners 'vibe coded' in the past month, and 62% used automation or agentic AI, with the average user employing 3.5 different models.

  • 2d ago

    Yampolskiy notes that AI models can already discover zero-day exploits, escape contained environments, and smuggle information using steganography, referencing the 'Mythos' model as an example.

  • 2d ago

    Doomberg views MicroStrategy's common stock as risky due to senior claims on Bitcoin by $15.2 billion in bonds and preferreds, arguing equity holders rely on perpetual share printing to service debt.

  • 3d ago

    Zico Kolter argues modern AI is conceptually simple, with core LLM training and RL code achievable in roughly 200-300 lines of Python.

  • 3d ago

    Ben Vinegar's startup modem focuses on product work and uses 'coding agent harnesses', while Armen's company Arendelle builds AI products like the email agent Leos and the coding agent Pi, which now operates as part of Arendelle.

  • 3d ago

    Cost pressure extends to AI tooling; for example, the Grapefruit code review tool's bill increased fivefold after switching from seat-based to usage-based pricing, reaching roughly $800 monthly for Ben's team.

  • 3d ago

    Vulnerability discovery harnesses like Warden, built on Claude's SDK, have found over one hundred security issues in codebases such as Sentry, demonstrating that AI-assisted security auditing is now widely effective despite often being dismissed as 'AI slop'.

  • 3d ago

    X.AI's strategic acquisition of Cursor for $60 billion was rationalized by Armen as a data-for-compute trade, where X.AI gains invaluable coding traces for model training while Cursor lacks its own GPU infrastructure.

  • 3d ago

    Armen asserts that open-weight AI models need access to high-quality coding traces to compete with large labs, leading Mario to share Pi's traces on Hugging Face, but creating such a dataset requires overcoming chicken-and-egg adoption challenges.

  • 3d ago

    GitHub's reliability is declining due to increased agent traffic, internal leadership issues, and data center migration pressures, prompting projects like HashiCorp's to consider leaving the platform and creating an opening for competitors.

  • 3d ago

    Ben argues the need to piggyback on GitHub for AI tool integration is decreasing as agents can now instrument tasks locally and tools like the dinosaur code review company operate across CI systems.

  • 3d ago

    The AI system generates a bespoke render of a luxury pool in the target backyard, calculates installation costs, and automatically mails a physical postcard with a QR code. This process leverages AI for image segmentation and image generation, as well as direct mail APIs with USPS integration.

  • 3d ago

    AI empowers solo business operators by reducing the need for traditional marketing, operations, or finance teams. Individuals can now handle these tasks independently or learn new skills efficiently through AI assistance, as Jake Woodhouse illustrates with WordPress troubleshooting using Claude.

  • 3d ago

    Ramp built an internal AI workspace called Glass to solve coordination and setup pain. It auto-configures with all company tools via SSO and includes a marketplace of over 350 reusable skills.

  • 3d ago

    Guillermo Rauch announced Vercel is open-sourcing a reference platform for cloud coding agents, aiming to help others build the type of internal systems companies like Stripe, Ramp, Spotify, and Block use.

  • 3d ago

    Nathan Day describes proof of place as a precursor to proof of personhood in his work on BTC Map. It involves mailing cryptographic proofs to verify physical access to a property and control of a private key.

  • 3d ago

    Nathan Day states the person NIP and required attestation NIP updates are nearly ready for release, aiming for a draft on NostrHub within a week of the recording.

  • 3d ago

    The proposed solution is using cryptography and open standards for decentralized identifiers (DIDs). This allows for cryptographic proof of authorship via digital signatures, shifting from probabilistic inference to mathematical certainty.

  • 3d ago

    Zero-knowledge proofs and selective disclosure enable privacy-preserving verification. A user can prove they are over 21 without revealing their birthdate, or prove they own a red hat without handing over the entire credential.

  • 3d ago

    The ideal credential flow involves an issuer crafting a signed credential bound to a public key, allowing the holder to generate a one-time, non-replayable ZK proof for a verifier. No personal information is stored or transmitted.

  • 3d ago

    The $285 million Drift Protocol hack involved North Korean proxies meeting protocol developers at conferences over months to gain access, representing a shift from technical attacks to social engineering at scale.

  • 3d ago

    Redbord believes privacy and security are not mutually exclusive, advocating for zero-knowledge proofs and permissioned visibility to allow lawful privacy while blocking illicit transactions.

  • 3d ago

    On prosecuting developers like Roman Storm, Redbord aligns with 'code is not crime' but argues criminal intent, such as conspiring with bad actors on darknet markets, justifies money laundering charges.

  • 3d ago

    Andreessen predicts the convergence of programmer, product manager, and designer roles into a new 'builder' role, empowered by AI to handle all aspects of product creation.

  • 4d ago

    Nathaniel Whittemore points to Q4 2025/Q1 2026 as an inflection point, driven by new models like Opus 4.5 and GPT 5.2, alongside transformative capabilities from Claude Code and Codex, leading to record frontier model releases.

  • 4d ago

    Leading AI companies reported immense revenue growth, with Claude Code reaching $2.5 billion annualized revenue, Cursor doubling to $2 billion, and Anthropic achieving a $19 billion run rate.

  • 4d ago

    Practitioner surveys show widespread AI usage, with 71% engaging in "vibe coding" and average users employing 3.5 models. The perceived value of AI shifted from time savings to increased output and new capabilities.

  • 5d ago

    Anthropic's Code with Claude event centered on agent harnesses and workflow tools rather than a major model release, reflecting the shift from model to ecosystem competition.

  • 5d ago

    Anthropic's Outcomes feature uses a separate grading agent with a user-defined rubric to automatically evaluate and iterate on agent outputs, improving document quality by 8.4% for Word and 10.1% for PowerPoint.

  • 5d ago

    Anthropic released a suite of 10 predefined financial service agents for Claude Finance, including pitch builders and market researchers, alongside a cookbook for customization.

  • 5d ago

    Claude Code creator Boris Cherny disavowed the term 'vibe coding', stating his company uses coordinated AI agents over Slack with automated testing and no manually written code.

  • 5d ago

    Anthropic will dissolve XAI as a separate company, integrating it as SpaceX AI, and will initially raise API rate limits and eliminate peak hour reductions for Claude Code users.

  • 5d ago

    Brian Elliott says Blitzy raised $200M at a $1.4B valuation and focuses on large-scale autonomous software development for enterprises, using multiple frontier models in orchestration to generate thousands of lines of code.

  • 5d ago

    Urban describes using mechanical calipers to measure broken parts and designing replacements in Blender, citing a custom-printed door handle that saved significant cost compared to a landlord's repair quote.

  • 6d ago

    Labor market data shows a crash in professional and business services job openings, multi-year lows, while demand for software engineers rises, signaling AI disruption of 'bullshit jobs' and a shift toward skilled technical roles.

  • 6d ago

    DK built the naming system prototype using Codex AI as a technical partner, allowing him to implement a complex economic system without deep prior Bitcoin programming expertise.

  • 6d ago

    Coder's new /goal feature allows an AI agent to work autonomously for days on a single objective, which Andrew Chen predicts will 10,000x token use.

  • 6d ago

    Stern used AI for editing and fact-checking her book but relied on a human editor for structural feedback, arguing AI lacked the ability to assess long-form narrative coherence.

  • 6d ago

    Git Workshop's web interface allows direct merging of pull requests, integrates NIP51 lists for starring repos, and uses encrypted NSEC notifications.

  • 6d ago

    Git over Nostr via ngit works for basic publication, but lacks CI/CD and mobile apps, making it unsuitable for production systems requiring low change costs.

  • 6d ago

    Marmot TS migrated addressable key packages from legacy Kind 443 to 30,443 to simplify SDK APIs and pave the way for multi-device support.

End of 7-day edition — 64 results