US Commerce Dept. greenlights Anthropic’s powerful Mythos 5 AI model for 100+ trusted companies and agencies. Still no update on consumer access to Fable 5.

"Technically based on the prospectus, Saylor can just let STRC fail and just pause dividends forever." "Stack real Bitcoin, not that fake shit." @MartyBent and (ODELLXYZ) on why real Bitcoin is the only answer.

"One-on-one tutoring works. One-on-one healthcare works. One-on-one finance works. AI is allowing companies to provide that at scale." - (APompliano)

OpenAI is leaning toward delaying its IPO until 2027 according to the New York Times. The reason: Sam Altman called anything below a $1 trillion valuation a "non-starter." When advisers presented two options, go public this year at a lower valuation or wait until 2027 for a shot at $1 trillion, Altman chose to wait. CFO Sarah Friar has reportedly been pushing for the delay, citing massive cash burn, compute infrastructure commitments, and the burden of public reporting. The decision isn't happening in a vacuum. SpaceX's IPO on June 11 was supposed to be the biggest tech listing in years. Shares priced at $135, rocketed to over $225, then gave back most of those gains. SpaceX now trades around $153, barely above its listing price. If SpaceX couldn't sustain momentum, OpenAI's team is right to wonder what happens when a company burning cash at their rate faces the same scrutiny. There's also the government angle. On the same day OpenAI announced GPT-5.6, it revealed the launch was limited because the US government asked them to. Following Trump's June 2nd executive order on AI assessments, OpenAI is now rolling out access "customer by customer" with government approval. That kind of regulatory uncertainty doesn't exactly make for a clean IPO story. The $1 trillion target tells you where Altman's head is. OpenAI's last private valuation was near $850 billion. Rather than accept a public market haircut, he'd rather bet that another year of revenue growth closes the gap. It's a bold move that assumes the AI hype cycle sustains through 2027 and that no competitor materially changes the landscape in the interim. SoftBank shares dropped over 12% on the news. When the most anticipated tech IPO in years would rather stay private than risk not hitting its number, it suggests the valuation gap between private and public markets remains wide and that the reckoning over AI capital expenditure returns is still ahead of us.

"There's a ton of companies right now laying people off and blaming AI. It's not because of AI. It's because they didn't do the layoffs in 2022 and 2023 when they should have." (APompliano) says AI is a scapegoat for companies that over-hired and mismanaged.

"AI-exposed companies are actually hiring more people and increasing wages faster. If I have people who are now more productive, I want as many of them as I can get." (APompliano) on why AI-exposed companies are hiring more, not less.

"He is probably the first person that went to a major corporation and said, you do not understand the plight of the average person. It is imperative that you spend time trying to learn this." (APompliano) argues there's more to Peter Thiel than the media tells you.

"I don't think it's gonna happen before midterms. And I also don't know if I want it to." @MartyBent and (ODELLXYZ) discuss the CLARITY Act and whether no bill is better than a bad bill.

OpenAI just dropped GPT-5.6 with a new three-tier system: Sol (flagship), Terra (balanced), and Luna (fast and affordable). But the release itself is almost secondary to how it's being released. This is a limited preview available to roughly 20 organizations. Not because OpenAI wants it that way, but because the US government asked them to. Following Trump's June 2nd executive order calling on federal agencies to assess new AI models before wide release, OpenAI previewed GPT-5.6 with the government ahead of launch. At their request, broad public access is being withheld until a framework is established around July 2nd. OpenAI is being transparent about their discomfort: "We don't believe this kind of government access process should become the long-term default. It keeps the best tools from users, developers, enterprises, cyber defenders, and global partners who need them." This comes directly after the US government issued an export control order against Anthropic's Claude Fable 5 and Mythos 5 over jailbreaks, forcing Anthropic to pull public access entirely. OpenAI is clearly trying to avoid that fate by coordinating proactively. On the technical side, Sol is OpenAI's most capable model yet, particularly in cybersecurity and coding. It introduces a "max" reasoning mode and an "ultra" mode deploying subagents for complex tasks. On Terminal-Bench 2.1, Sol Ultra scores 91.91%, edging out Claude Mythos 5 at 88%. The pricing is worth noting. Sol matches GPT-5.5 at $5/$30 per million tokens. But compare that globally: DeepSeek v4 Flash runs at $0.14/$0.28, Xiaomi's MiMo at $0.10/$0.30. OpenAI's cheapest option is still mid-priced by global standards. Chinese labs are shipping frontier capability at a fraction of the cost. The bigger story isn't the model. It's the precedent. We now have a framework where the US government reviews AI releases before they reach the public. OpenAI calls it temporary. But temporary government powers have a way of becoming permanent. The executive order gives agencies 30 days to build a "repeatable process for future model releases." That's a regulatory framework being built in real time, and the question is whether it makes anyone safer or just ensures the most capable models are available everywhere except to the citizens of the country that built them.

"The first Stratum V2 block was mined by Demand Pool. Many people have been speculating whether or not this would ever be possible. It happened today." @MartyBent and (ODELLXYZ) on the milestone years in the making.

Senators are calling for a formal investigation into Polymarket's advertising practices, the CFTC reportedly already has an ongoing investigation, and a consumer protection group has filed a lawsuit. All of it stems from the Wall Street Journal's investigation last week. WSJ found that Polymarket paid mostly college-aged creators $2,000-$3,000 per month to post videos of fabricated winning bets. The videos weren't filmed on the real platform. They were filmed on dummy websites designed to look identical, with URLs like poiymarket .com. Across 1,105 TikTok videos, the depicted wagers totaled $1.9 million. None were real. One example: a college student appeared to win $100,000 betting $1,000 that Trump would say "McDonald's" publicly. In reality, every account that placed that bet on the actual platform lost. Of the 118 videos depicting wins totaling $900,000, the same bets placed for real would have resulted in over $166,000 in losses. Creators were explicitly told not to disclose their paid relationship with Polymarket, violating FTC disclosure requirements. The campaign was run through a marketing contractor who assembled what WSJ called a "social media army" of clippers to amplify the content. This comes on top of Polymarket's $1.4 million CFTC fine in 2022, a Google employee charged with insider trading on the platform, and accounts flagged by 60 Minutes for suspiciously accurate bets on US military operations.

"I'm not saying I believe it, I think it's plausible, but that Strategy is the strategic reserve." @MartyBent on whether Strategy is the US government's shadow Bitcoin accumulation play.

Most people are drowning in disconnected headlines. The Bitcoin Brief connects the dots across Bitcoin, macro, energy, AI, freedom tech, and sovereignty so you can see where power is moving before the crowd does. Sign up for free:

TFTC – Truth for the Commoner

Subscribe to Bitcoin Brief

Subscribe to Bitcoin Brief

"The biggest piece here is the 800,000 coin gorilla in the room with an orange tie being MicroStrategy. If Saylor checks his ego he can stop the bleeding in a big way." @MartyBent and (ODELLXYZ) discuss whether Saylor's next move defines this bear market.

People who use AI in the most automated, agentic way are the MOST optimistic about what it means for their work, anticipating positive impacts on pay, job security, and meaning. The people closest to the automation are the least afraid of it. That's one of the key findings from Anthropic's latest Economic Index report, which now samples Claude usage at the hourly level. The patterns reveal just how deeply AI has embedded itself into daily life. Personal usage spikes from 35% on weekdays to nearly 50% on weekends. During the week it's marketing copy, business emails, and slide decks. On weekends people shift to emotional support, medical questions, and investment advice. On nights and weekends, the work that does happen skews heavily toward higher-wage occupations. The knowledge workers never really clock out. The daily rhythm is almost poetic. People ask Claude for news at 7am. Business emails peak at 10-11am. Recipe requests spike 2.3x at 6pm. Sleep advice peaks at 3am. Tax-related queries surged 8x right before the April 15 filing deadline. AI usage is a mirror of human life. Conversations mapped to higher-wage occupations consume significantly more tokens. And crucially, in these higher-value conversations, both Claude AND the human are doing more work. More output per turn, more turns, more extended thinking. This looks like augmentation, not displacement. Across countries, conversations related to starting a business peak on Saturday and Sunday. People are using their free time to build. AI agent design, quant trading, and gaming projects all surge on weekends while backend architecture and API debugging drop off. This is what the early stages of an economic transformation look like. Not a sudden displacement event but a gradual deepening. AI is following the cadence of human life.

One man is worth more than the top billionaires of China, Hong Kong, Japan, and South Korea combined. Elon Musk: $957B

Japan just posted the largest crude oil reserve drawdown in the country's history. Inventories collapsed from ~355 million barrels to below 280 million in a matter of weeks, blowing well past the 5-year range.

"What are the odds Bitcoin becomes the global currency? If the odds are 15, 20%, that's probably the $40 trillion to $100 trillion range. That's $2 to $5 million per Bitcoin. A 30 to 100x from here." - @Nik Bhatia

"AI has the ability to lift humanity up. Bitcoin allows people to preserve any wealth they've created. You need Bitcoin no matter how good or how bad the world is going." - @Nik Bhatia

Microsoft just announced its third Xbox price hike in 14 months. Starting August 1, the Xbox Series S jumps $100 to $500, the Series X starts at $750, and the 2TB model is being discontinued entirely. The reason is worth paying attention to. Console storage and memory prices have increased by more than 2.5x since last fall, and Microsoft expects another doubling by fall 2027. By the 2027 holiday season they expect to be paying over 5x what they paid two years earlier for the same components. This is not a tariff story. This is an AI story. AI data centers are consuming enormous quantities of memory chips. High Bandwidth Memory for AI accelerators now takes up 23% of all DRAM wafer production. Enterprise SSDs for AI workloads are absorbing NAND supply that used to go to consumer products. Manufacturers are reallocating production toward these higher-margin AI products because that's where the money is. DRAM contract prices surged 58 to 63% in Q2 2026 alone. NAND flash jumped 70 to 75% in the same quarter. The tech press has started calling it "RAMmageddon." The result is that every consumer electronics product that needs memory or storage is getting squeezed. Microsoft is raising Xbox prices. Apple's Tim Cook said iPhone price hikes are "unavoidable." Dell already raised PC prices. Ford flagged DRAM shortages affecting vehicle pricing. A coalition of retailers, media companies, and medical supply companies wrote to the White House warning about "an urgent imbalance in the market for memory chips." The AI boom is real. But the chips going into AI data centers are the same chips that go into your gaming console, your phone, your laptop, and your car. When the biggest companies in the world are willing to pay any price for memory to power their AI infrastructure, everyone else gets crowded out. Your Xbox getting $100 more expensive is one of the first visible signs. It will not be the last.