🤖 It’s Not Too Dangerous for the General Public

🔐 It’s Not Too Dangerous for the General Public

🇺🇸 Will AI Become a Public Utility?

📍 6 AI Stock Sectors

🛠️ 5 Trending Tools

🥪 4 Brief Bites

💰 3 Funding Frontlines

💼 4 Job Opportunities

Anthropic recently announced “Project Glasswing,” a new initiative to secure critical software for the AI era. This new initiative was formed because of the cybersecurity risks observed with “Claude Mythos Preview,” the world’s most capable frontier AI model for agentic coding. It autonomously found thousands of high-severity software vulnerabilities across every major operating system and web browser. As a result, Anthropic chose not to make it publicly available. Well, that ship has sailed.

Anthropic released “Claude Fable 5,” a Mythos-class model with added safety guardrails that make it suitable for everyday use. For example, it’s equipped with real-time classifiers that instantly detect potential misuse, rerouting sensitive prompts involving cybersecurity exploitation and biological weapons to the more restricted “Claude Opus 4.8.”

It achieved 80.3% accuracy on SWE-Bench Pro, which evaluates AI agents on real-world, long-horizon software engineering tasks sourced directly from GitHub. For context, that’s the highest recorded accuracy score ever. It also achieved 46.1% accuracy on BioMysteryBench, solving nearly half of the 23 biological research mysteries that stumped 5 expert virologists.

It’s priced at $10.00 per million input tokens and $50.00 per million output tokens. That’s 2x the cost of “Claude Opus 4.8.” It’s available on Pro, Max, Team, and seat-based Enterprise plans at no extra cost through June 22nd, 2026. After that, access will require usage credits. So, make the most of it over the next 12 days unless your wallet doesn’t mind the bill.

Anthropic raised a $65 billion Series H funding round at a $965 billion post-money valuation, becoming Silicon Valley’s highest-valued frontier AI firm. Anthropic also expects to achieve its first profitable quarter ever, with projected quarterly revenue of $10.9 billion for Q2 FY2026. The only widespread grievance with Anthropic is cost, especially regarding how quickly you burn through token limits and usage credits.

The Trump Administration and OpenAI CEO Sam Altman are in ongoing talks about the U.S. government taking a direct equity stake in OpenAI. Under this sovereign-wealth-style investment vehicle, OpenAI would voluntarily donate up to a 5% ownership stake to seed a “Public Wealth Fund (PWF).” In simple terms, it means the PWF would own up to 5% of OpenAI. This would make every U.S. citizen a literal shareholder with a stake in “AI-driven economic growth.”

The PWF would hold U.S. government equity stakes in Big Tech companies, generating investment returns and portfolio yields over time that are redistributed directly to U.S. citizens through a “universal basic dividend,” rather than having the U.S. government spend this wealth to subsidize bureaucratic services.

U.S. President Donald Trump explained, “There’s something very interesting about it, where it almost becomes a partnership with the American public, and that would be a beautiful thing. It would make ‘em rich.” It broadly aligns with Trump’s publicly expressed interest in U.S. government ownership of Big Tech companies. Notably, taking a direct 9.9% equity stake in Intel by purchasing $8.9 billion worth of primary shares of common stock.

During a WSJ TechLive event, OpenAI CFO Sarah Friar accidentally stated that OpenAI supports “federal backstops” to finance AI infrastructure buildouts. After a wave of backlash, OpenAI went into crisis mode and clarified the U.S. government shouldn’t pick winners or losers and taxpayers shouldn’t bail out bad business decisions. The PWF adds another layer of complexity by deepening ties between OpenAI and the U.S. government.

Former AI & Crypto Czar David Sacks came out against this sovereign-wealth-style investment vehicle, warning it could blur the line between public governance and private enterprise, destroying regulatory neutrality with inherent conflicts of interest. He believes the only influential role the U.S. government should play is streamlining the regulatory approval process needed to build AI infrastructure.

Sandstone raised a $30M Series A for in-house legal AI.

PointFive closed a $60M Series B to control AI cloud costs.

PhysicsX landed a $300M Series C to scale industrial physics AI.

Scale AI (New York, NY): Agentic AI Builder Intern, Summer 2026

NVDIA (Santa Clara, CA): Formal Verification Engineer, Entry-Level

Datadog (Dublin, IE): Deal Desk Quoting Price Analyst, Mid-Level

Redis (San Francisco, CA): Sr. Applied AI Engineer, Senior-Level