← All briefings
📰

subtl daily briefing

Share𝕏in
AllAICryptoSaaSFinanceStartups

Good morning, founders and investors. OpenAI dropped GPT-5.5 with autonomous agentic capabilities today, and Anthropic just crossed a $1 trillion valuation — making this one of the most consequential weeks in AI history. Meanwhile, a $293M DeFi hack, a soldier charged for Polymarket insider trading, and Meta cutting 8,000 jobs to fund AI infrastructure are all reshaping the landscape. Let's get into it.

In today's briefing

  • 1.GPT-5.5 Launches, Anthropic Hits $1T
  • 2.Meta Cuts 8,000 Jobs to Fund AI Bet
  • 3.Kelp DAO $293M Hack Rocks DeFi
  • 4.Bitcoin Surges 24% Amid U.S.-Iran War
  • 5.The 'Gen Marketer' — AI's New Growth Hire
  • ⚡Quick hits on other news
Latest Developments
AI

🤖GPT-5.5 Goes Live and Anthropic Crosses $1 Trillion Valuation

The Rundown: OpenAI launched GPT-5.5 with advanced agentic capabilities and a 1M token context window, while rival Anthropic surpassed OpenAI's own valuation by crossing $1 trillion on Forge Global.

The details:

  • ●GPT-5.5 launches with 82.7% on Terminal-Bench 2.0, $5/M input tokens API pricing, and a 1M token context window — available in ChatGPT and Codex now.
  • ●OpenAI Workspace Agents automate Slack, email, spreadsheets, and calendars; free until May 6, 2026 on Business, Enterprise, Edu, and Teachers plans.
  • ●Anthropic surpassed OpenAI in valuation at $1 trillion on Forge Global, driven by Claude Code adoption and scarcity of available shares.
  • ●DeepSeek V4-Pro launched simultaneously with 1.6 trillion parameters and 1M token context, confirmed compatible with both Nvidia and Huawei chips — and in talks to raise at a $20B valuation backed by Tencent and Alibaba.
Why it matters: For founders, the $5/M token pricing for GPT-5.5 and the proliferation of capable open-weight models like Qwen3.6-27B (which beats Alibaba's own 397B model on coding benchmarks while running on 18GB VRAM) means AI infrastructure costs are compressing fast. One real-world data point buried in today's newsletters: a Claude integration replicated 95% of a SaaS vendor's AI features at 15% of the token cost — and the customer slashed their renewal price by 45%. That's the pricing pressure every SaaS founder needs to be stress-testing against right now.

📰 Source: TLDR, AlphaSignal, Techpresso

Share𝕏in
AI

📉Meta Cuts 8,000 Jobs and Cancels 6,000 Open Roles to Bet on AI Tokens Over People

The Rundown: Meta is eliminating roughly 8,000 jobs and redirecting resources toward up to $135 billion in AI infrastructure spending, embodying a broader corporate shift from human headcount to AI compute.

The details:

  • ●Meta is cutting 8,000 jobs and cancelling 6,000 open roles, with layoffs effective May 20, to free up funds for AI infrastructure investment.
  • ●Google reports 75% of its code is now AI-written, and its API saw 60% token growth in Q1 driven by agentic AI adoption — validating Meta's strategic logic.
  • ●S&P 500 companies have reported only $300M in total AI productivity gains even as Anthropic earns roughly that amount every 1.5 days in token sales — raising real questions about where the value is accruing.
  • ●Microsoft is also offering buyouts to 7% of its U.S. workforce as part of a similar AI-investment pivot.
Why it matters: The Meta and Microsoft moves signal that the 'tokens vs. people' trade-off is no longer theoretical — it's a live reallocation happening at the largest companies in the world. For founders building SaaS on top of human workflows, this is both a threat and an opportunity: the enterprises that are your customers are actively asking whether AI can replace what you help their employees do. The dangerous data point here is that the productivity gains aren't yet showing up at the macro level ($300M across the S&P 500), which suggests we're in an investment phase, not a returns phase — but the companies who nail the transition early will have a massive structural cost advantage.

📰 Source: Techpresso, The Breakdown, MIT Technology Review

Share𝕏in
⚡

Everything else in the news today

A U.S. soldier faces 5 felony charges for using classified intel on Maduro's capture to earn $400K on Polymarket — the first-ever insider trading case involving a prediction market.→
NY AG Letitia James sued Coinbase and Gemini for operating unlicensed prediction markets, while Wisconsin separately sued Kalshi, Robinhood, and Coinbase for illegal gambling.→
Project Eleven awarded quantum cryptographer Giancarlo Lelli 1 BTC for cracking a 15-bit elliptic curve key, raising alarms about Bitcoin's post-quantum security timeline.→
Z.ai released GLM-5.1, a 754B parameter open-weights MoE model that can autonomously loop through coding tasks for up to 8 hours — priced at $1.40/$4.40 per million tokens under MIT license.→
The Bitwarden CLI npm package was briefly compromised via a malicious credential stealer targeting npm tokens, GitHub tokens, SSH keys, and cloud credentials.→
Researchers demonstrated that the Zealot AI multi-agent system can autonomously breach a GCP environment — from network recon to SSRF exploitation to BigQuery data exfiltration — without human intervention.→
DoorDash went live on Tempo for stablecoin-powered payouts across 40+ countries, joining Stripe on the same Stripe/Paradigm-incubated chain processing $10B+ annualized volume.→
Anthropic's Mythos cybersecurity model surfaced thousands of vulnerabilities across major OSes and browsers, prompting Australia to engage Anthropic over concerns it could accelerate sophisticated attacks.→
Band emerged from stealth with $17M to build a universal orchestration layer enabling AI agents across different frameworks and clouds to discover and collaborate in real time.→
AI systems now apply a 'bland tax' to generic content, filtering it out of AI-powered search results — making brand distinctiveness a literal algorithmic requirement, not just a creative preference.→
Cognition AI (maker of Devin) is in talks to raise at a $25B valuation.→
Kubernetes v1.36 shipped with 70 enhancements including fine-grained kubelet API authorization, while deprecating the vulnerable externalIPs field.→
Tokenmaxxing — employees at Meta, Microsoft, and Salesforce wastefully burning AI tokens for internal leaderboard rankings — is causing financial waste and system outages.→
A worm-like npm supply-chain attack targeting Namastex Labs packages is stealing developer credentials and self-propagating to PyPI — requiring immediate secret rotation if you use their packages.→
ChatGPT Images 2.0 is now live across all major Figma products including Design, FigJam, Slides, and Weave.→
📰Today⚡Feed📡Signals💰Capital