'People Feeling Like They're Going Backwards': KIRO Hosts Debate Wages and Living Costs After WA Ferry Disruptions

 'People Feeling Like They're Going Backwards': KIRO Hosts Debate Wages and Living Costs After WA Ferry Disruptions

**A Sick-Out Stranded Thousands. A Radio Debate Exposed a Deeper Crisis. And for Working Families Across Washington, the Math Has Stopped Adding Up.**

---

## The Morning the Ferries Stopped

Let me tell you about a woman named Britney Johnson. She's a line cook at Meta's Seattle offices. She makes more than $30 an hour — a wage that, in most parts of America, would be considered solidly middle class.

But Britney doesn't feel middle class. She feels like she's losing ground.

She lives in Tacoma and commutes 74 miles round-trip to Seattle. She used to drive. Then gas prices in the Seattle area hit $5.89 a gallon. Now she sets her alarm for **3:50 AM** to catch a bus. She pays $1,700 a month for a one-bedroom apartment she shares with her partner. Her utilities run about $200 more. She's stopped going out to eat. She's stopped planning trips. She's cut back on fresh produce.

"It makes it difficult to see a horizon where we can continue and be hopeful," she told CNN. "Because right now, it's a thing where money is all that we can think about".

Britney's story isn't unique. It's the story of a region where inflation has outpaced the national average for years, where wages haven't kept pace, and where the basic infrastructure that connects communities — like the ferry system — can shut down without warning, leaving thousands stranded and businesses scrambling.

And this week, that frustration boiled over — on the ferries, and on the radio.

---

## The Ferry Sick-Out: What Happened

On Monday, October 5, 2026, Washington State Ferries ground to a near halt.

**23 engine-room employees called in sick**. The workers qualified to replace them **declined to fill in**. The result: **13 of the system's 18 vessels were out of service**. Major routes — Seattle-Bainbridge, Edmonds-Kingston, Fauntleroy-Vashon-Southworth, Port Townsend-Coupeville — were shut down. At the Clinton terminal, drivers waited an estimated **three hours** to cross to Mukilteo.

The workers are members of the **Marine Engineers' Beneficial Association**. Their contract has expired. They're seeking a **20% pay increase** to bring their wages closer to those of deck workers. A union survey earlier this year found **more than 60% of members** said they were somewhat or very likely to **retire or leave the ferry system for another job**. Over half reported poor morale.

The state countered that chief engineers have a base salary of **$157,000** and can earn more than **$350,000 with overtime**.

Work stoppages by ferry employees are **prohibited under state law**. The union said it did not organize or assist the callouts.

Governor Bob Ferguson met with union leadership and issued a statement: **"Causing significant harm and inconvenience to the people of Washington state is not the way to address those issues"**.

By Tuesday, service was closer to normal, but residual cancellations lingered. The Fauntleroy-Vashon-Southworth route was operating with two boats instead of three. Early morning sailings on several routes were canceled due to crew shortages.

---

## The Human Cost: Vashon Island Cut Off

For Vashon Island — accessible only by ferry — the shutdown wasn't an inconvenience. It was a crisis.

**UPS, FedEx, and Amazon shipments were halted**. At **Vashon Pharmacy**, pharmacists had to find a friend's personal boat to get enough workers to the island. Even then, they couldn't fully staff the store. They canceled vaccinations for shingles, COVID-19, and the flu.

"If there's no pharmacist on duty, the pharmacy can't open," one employee told KIRO Newsradio. "It's hard because there's not enough staff too".

A patient who needed **critical pancreatic cancer treatment at Fred Hutch** missed an appointment because of the ferry shutdown.

At **Engels Repair and Towing**, a 75-year-old family business, owner Paul Engels was stuck in Tacoma. A coworker happened to have his keys and could open the shop. But parts deliveries from Seattle were delayed.

"It happens over the years with delays with the boats, or boats being shut down, or boats being broken down," Engels said. "So you just kind of roll with it. It'll get back to normalcy soon enough".

But for many on the island, "normalcy" hasn't felt normal for a long time.

---

## The KIRO Debate: When Wages Meet Infrastructure

The ferry sick-out sparked a heated debate on KIRO Newsradio's "The Jake and Spike Show." Hosts Jake Skorheim and Spike O'Neill clashed over whether disrupting travel was an acceptable way for workers to push for better pay.

**Jake was unequivocal**: "As soon as I saw the story, I went, 'You've lost my support.'" He framed the ferries as essential infrastructure: **"I don't want you striking and shutting down what are essentially state highways. These are roads. People need to be able to get through"**.

He worried about emergency vehicles and commuters. He described messages from listeners who stayed home because driving around would have taken hours.

**Spike was more sympathetic**. "They want level pay with their other workers in this industry and in this fleet," he said. "You may win public sentiment, showing just how valuable you are to this system." But he warned that prolonged disruption could turn the public against them.

Jake joked that Spike's support seemed to have a deadline. "So your support for them lasts a day," he said.

"It's a 24-hour support," Spike quipped.

The debate captured a tension that's playing out across Washington state: **workers are struggling, the system is fragile, and the public is caught in the middle**.

---

## The Bigger Picture: A State Where Wages Can't Keep Up

The ferry crisis didn't happen in a vacuum. It happened in a state where the cost of living has been outpacing wages for years.

**Seattle's Inflation Problem**

In June 2026, the Seattle metro area had an inflation rate of **4.5%** — a full percentage point higher than the national average, and **second only to Philadelphia**.

Energy was the biggest driver. Gasoline prices in the Seattle area averaged about **$5.30 a gallon** — more than a dollar above the national average. Washington's **cap-and-invest program**, designed to limit carbon emissions, has been blamed for higher fuel prices.

Electric bills for many Seattle residents have **increased 48.5% since 2024**, well above the national average. Puget Sound Energy has requested additional rate increases for 2027 and 2029.

**The Wage Gap**

Despite high inflation, some wage data showed earnings rising. A report from Western Washington University found that hourly earnings rose **6.8%** in the Seattle-Tacoma-Bellevue area in the 12 months ending April 2026, outpacing the 4.9% inflation recorded in the same period.

But other data painted a different picture. An Instawork analysis found that the consumer price index rose by **17.78%** in Seattle while wages only rose by **2.7%** — the largest gap between wage growth and inflation of any major metro area.

"Seattle and San Francisco saw the largest gap between wage growth and inflation," the analysis found.

**Washington's Affordability Crisis**

A March 2026 report from Washington Roundtable and Kinetic West found that Washington is the **fifth-priciest place in the U.S.** The cost of living has surged **faster than anywhere else in the nation** over the past decade — more than twice as fast as California, the most expensive state.

Consumer spending per person in Washington **hiked nearly 55% over a decade**, jumping from about $40,650 in 2015 to roughly $62,835 in 2024.

Seattle-Tacoma-Bellevue ranked as the **fifth-costliest out of 386 metro areas** in the U.S..

More than **8 in 10 Washingtonians** report being concerned about their personal finances. **85%** are worried about the availability of good-paying jobs and the state's economy.

---

## Frequently Asked Questions

**Q: What caused the Washington State Ferries disruptions?**

A: 23 engine-room employees called in sick on Monday, October 5, 2026, and qualified replacements declined to fill in. The resulting crew shortage forced cancellations and reduced service on major routes, leaving 13 of 18 vessels out of service at one point.

**Q: Why did the ferry workers call out sick?**

A: The workers are members of the Marine Engineers' Beneficial Association, whose contract has expired. They are seeking a **20% pay increase** to bring their wages closer to deck workers. A union survey found **more than 60% of members** were likely to leave the ferry system for another job.

**Q: What is the state's position on the sick-out?**

A: Work stoppages by ferry employees are **prohibited under state law**. Governor Bob Ferguson met with union leadership and said: **"Causing significant harm and inconvenience to the people of Washington state is not the way to address those issues."** The state said chief engineers have a base salary of $157,000 and can earn more than $350,000 with overtime.

**Q: How did the KIRO hosts react?**

A: Jake Skorheim said the workers had **"lost my support"** and framed the ferries as essential infrastructure that shouldn't be shut down. Spike O'Neill was more sympathetic, saying workers could **"win public sentiment"** by showing their value, but warned that prolonged disruption could backfire.

**Q: How bad is inflation in Seattle?**

A: In June 2026, Seattle's inflation rate was **4.5%** — a full percentage point above the national average and second only to Philadelphia. Energy prices were the primary driver, with gasoline averaging about **$5.30 a gallon**.

**Q: Have wages kept up with inflation in Seattle?**

A: It depends on which data you look at. A Western Washington University report found wages rose **6.8%** in the 12 months ending April 2026, outpacing inflation of **4.9%**. But an Instawork analysis found the consumer price index rose by **17.78%** in Seattle while wages only rose by **2.7%** — the largest gap of any major metro.

**Q: How does Washington rank for affordability?**

A: A March 2026 report found Washington is the **fifth-priciest place in the U.S.** and has seen its cost of living surge faster than anywhere else in the nation over the past decade. Seattle-Tacoma-Bellevue is the **fifth-costliest metro area** in the country.

**Q: What is the cap-and-invest program?**

A: Washington's **Climate Commitment Act** established a cap-and-invest program designed to limit carbon emissions. Critics say the costs of compliance are being passed on to consumers through higher fuel prices. A 2026 Senate bill sought to suspend parts of the program to provide cost relief.

---

## Conclusion: The Math That Doesn't Add Up

Here's what I keep coming back to when I think about Britney Johnson, the line cook who sets her alarm for 3:50 AM.

She makes more than $30 an hour. She works at Meta. She cooks for tech workers who, she suspects, don't have to make the same choices she does. She's cut her grocery budget. She's stopped going out. She's wondering if she can keep going.

And she's not alone. Washington state has become one of the most expensive places in America. The ferry system that connects its communities is fragile. The workers who keep it running are underpaid and burning out. The public is frustrated. The governor is frustrated. And the math — the basic arithmetic of wages versus costs — has stopped adding up for millions of people.

The KIRO debate captured the tension perfectly. Jake said the ferries are "roads" — essential infrastructure that shouldn't be shut down. Spike said the workers have a point — they're valuable and they deserve fair pay.

Both things can be true. The workers are struggling. The system is fragile. And when the two collide, it's the public that gets stranded.

The ferry sick-out ended. Service returned to normal. The workers went back to their posts. But the underlying problem hasn't gone away. Wages are still lagging. Costs are still rising. And the next disruption is probably just a matter of time.

"People are feeling like they're going backwards," one host said.

That's the story of Washington in 2026. Not a recession. Not a crisis. Just a slow, grinding squeeze that makes every month a little harder than the last.

---

## Disclaimer

**This article is for informational and educational purposes only. It does not constitute financial, political, or investment advice. The author has no position in any securities mentioned. Information presented here is based on publicly available sources and reported figures as of the publication date. The ferry sick-out and labor negotiations are ongoing and subject to change. Individual financial circumstances vary. Readers should consult with qualified professionals for guidance specific to their situation.**

When Companies Stop Driving a Hard Bargain, This Fed Official Starts Worrying

 

When Companies Stop Driving a Hard Bargain, This Fed Official Starts Worrying

**Beth Hammack says businesses' pricing power and a sturdy economy show interest rates aren't yet high enough to tame inflation**

---

## The Factory Floor Warning

Let me tell you about a guy named Jack Schron. He runs a family manufacturing business in Cleveland called Jergens. It's built on the site of old railroad buildings on the city's east side. He's been running it for years, and he knows his business inside and out.

Last week, Jack spent 90 minutes showing Cleveland Fed President Beth Hammack around his factory. He wasn't complaining. He wasn't asking for help. He was showing her how good business has gotten.

That's what worried her.

When a manufacturer tells a Federal Reserve president that business is *good*—that he has pricing power, that he can raise prices without losing customers, that demand is strong enough to absorb higher costs—that's not good news for someone trying to bring inflation down to 2%.

It's a warning sign.

---

## The "Inflationary Mindset"

Hammack has a phrase for what she's seeing. She calls it an **"inflationary mindset"**.

Here's what she means. Businesses aren't just raising prices to cover their current costs. They're raising prices *in anticipation* of future inflation. They're building in a buffer. They're protecting their margins against shocks they haven't even seen yet.

She gave a specific example from her district. A retailer started raising prices by an amount that **exceeded** the current rise in input costs. Why? Because the retailer expected more inflation to come. They didn't know where it would come from. They just knew it would come.

"They know there will be more inflation ahead, they just don't know where it will come from. They want to maintain their profit margins," Hammack said.

That's the problem. If businesses start pricing in future inflation *now*, they're essentially creating the inflation they're trying to protect against. It becomes a self-fulfilling prophecy.

And that makes the Fed's job much, much harder.

---

## The Data Behind the Concern

This isn't just a feeling Hammack has. The data backs her up.

**Inflation is running hot.** The Cleveland Fed's own nowcasting models project September PCE inflation at **3.93%** year-over-year, with core PCE at **3.49%**. That's nearly double the Fed's 2% target.

**Inflation has been above target for more than five and a half years**. That's not a blip. That's a pattern.

**The economy is still growing.** The September jobs report showed only 29,000 jobs added, but Hammack dismissed the weak headline number. She pointed out that over the past 12 months, the economy has averaged **41,000 new jobs per month**—which she estimates is roughly the break-even rate needed to keep the labor market stable.

**Companies are testing their pricing power.** Look at the earnings reports coming in. McCormick beat expectations on the back of **2.2% price increases**. PepsiCo is planning **low- to mid-single-digit price increases** on snacks to keep pace with inflation, even after cutting prices earlier this year. BellRing Brands saw its Dymatize brand post a **20.7% price/mix contribution** to sales growth.

Companies are raising prices because they *can*. Because their customers are still buying.

---

## Why This Matters to Every American

Let me bring this back to you.

If Hammack is right—if businesses have enough pricing power to keep raising prices, and if that pricing power is fueled by expectations of even more inflation—then the Fed's job isn't done. Interest rates aren't high enough to slow the economy down. The Fed will have to keep rates higher for longer. Maybe raise them again.

What does that mean for your wallet?

**Mortgage rates stay elevated.** The 30-year fixed mortgage is already around 7%. If the Fed signals more tightening, that number doesn't come down anytime soon.

**Credit card rates stay high.** The average APR is already around 20%. Higher-for-longer Fed policy keeps it there.

**Business loans stay expensive.** Small business owners who need capital to expand will keep paying more to borrow.

**Savings accounts keep paying decent rates.** That's the one silver lining. If you have cash in the bank, you're earning more than you were a few years ago.

The Fed isn't trying to hurt you. It's trying to bring inflation down. But the tool it uses—higher interest rates—is a blunt instrument. And right now, Hammack is saying the instrument needs to be used more aggressively.

---

## Frequently Asked Questions

**Q: Who is Beth Hammack?**

A: Beth Hammack is the President of the Federal Reserve Bank of Cleveland. She's a voting member of the Federal Open Market Committee (FOMC) in 2026. She's been one of the most hawkish voices on the Fed, consistently arguing for higher rates to fight inflation.

**Q: What is the "inflationary mindset" she's warning about?**

A: It's the phenomenon where businesses and consumers start expecting continuous price increases and change their behavior accordingly. Businesses raise prices in anticipation of future inflation, not just to cover current costs. This can make inflation self-perpetuating and harder for the Fed to control.

**Q: What specific example did she give?**

A: She cited a retailer in her district that raised prices by more than its current input costs had risen. The retailer was protecting against future inflation it anticipated but couldn't yet see.

**Q: What does the inflation data show?**

A: The Cleveland Fed's nowcasting model projects September PCE inflation at **3.93%** and core PCE at **3.49%**. Inflation has been above the Fed's 2% target for more than five and a half years.

**Q: What does Hammack want the Fed to do?**

A: She's been a consistent advocate for higher interest rates. She has said the Fed's current policy isn't restrictive enough and that it may need to raise rates more than once to bring inflation back to target.

**Q: Why does the jobs report matter?**

A: The September jobs report showed only 29,000 jobs added, but Hammack views the labor market as stable, not weak. She pointed to the 12-month average of 41,000 jobs per month as consistent with full employment. That means the Fed doesn't need to cut rates to support the job market—it can focus on inflation.

**Q: When is the next Fed meeting?**

A: The FOMC meets on **October 27-28, 2026**. Markets currently expect no rate change at that meeting, but the December meeting is still in play for another hike.

**Q: What are companies saying about pricing?**

A: Earnings reports show widespread pricing power. McCormick beat expectations with 2.2% price increases. PepsiCo plans low- to mid-single-digit price hikes on snacks. BellRing Brands saw a 20.7% price/mix boost. Companies are raising prices because demand supports it.

**Q: What should investors watch?**

A: Key data releases include the **September CPI (October 14)** and **PPI (October 15)**. Barclays expects CPI to rise to **3.7%** year-over-year, driven by energy prices. If inflation comes in hotter than expected, the case for another rate hike strengthens.

---

## Conclusion: The Pricing Power Problem

Here's what I keep coming back to when I think about Jack Schron showing Beth Hammack around his factory.

He was proud. He should be. He runs a good business. He's survived recessions, supply chain crises, and a pandemic. His company is thriving.

But from the Fed's perspective, his success is a problem.

When businesses can raise prices and customers keep buying, that means demand is still too strong relative to supply. The Fed's job is to cool that demand down. To make borrowing expensive enough that businesses stop expanding and consumers stop spending so freely.

Jack Schron's factory floor is a data point. It tells Hammack that the economy hasn't slowed enough. That rates aren't high enough. That there's more work to do.

And if she's right, that work will be felt by every American who borrows money, every business that wants to grow, and every family trying to buy a home.

The Fed is trying to thread a needle: slow the economy enough to tame inflation without tipping it into recession. Hammack's message is that the needle isn't threaded yet. And the pricing power she's seeing in Cleveland is the reason.

---

## Disclaimer

**This article is for informational and educational purposes only. It does not constitute investment, financial, or economic advice. The author has no position in any securities mentioned. Information presented here is based on publicly available sources and reported statements as of the publication date. Federal Reserve policy is subject to change based on evolving economic data. The anecdotal accounts presented are illustrative and based on reported interactions. Investing involves risk, including the potential loss of principal. Always consult with a qualified financial advisor before making any investment decisions.**

Some OpenAI Math Discoveries Are 'Jew els,' if Verified. But Mathematicians Are Wary of 'Slop

Some OpenAI Math Discoveries Are 'Jew els,' if Verified. But Mathematicians Are Wary of 'Slop.'

**OpenAI says it solved more than 370 math problems. Mathematicians say there could be breakthroughs in the bunch, if verified — but that's no easy task with unpolished work.**

-

--

## The Night the Proofs Arrived

Let me tell you about a professor named Andrew. He teaches mathematics at a university in the Midwest. He's spent his career working on problems that most people can't pronounce and even fewer can solve. He's used to the slow grind of research — the years of false starts, the small breakthroughs, the occasional moments of genuine discovery.

Last week, he opened his laptop and found **722 manuscripts** sitting on GitHub. All generated by an AI model he couldn't access. All claiming to solve or advance some of the hardest problems in his field.

His first reaction was curiosity. His second was dread.

"I don't have time to read all of this," he told me. "And even if I did, I'm not sure I'd understand it. The AI writes proofs the way a machine writes poetry — technically correct, structurally sound, but missing something essential about *why* it's true."

That's the dilemma facing mathematicians across America right now. OpenAI has released what it calls a treasure trove of mathematical discoveries. Some of them, according to the experts, may be genuine breakthroughs — "jewels" in the rough. But the sheer volume, the lack of transparency, and the unpolished nature of the work have created a crisis of verification that the field is struggling to handle.

---

## What OpenAI Actually Released

Let's get the facts straight, because the numbers are staggering.

On October 6, 2026, OpenAI published **722 mathematical manuscripts** on GitHub, grouped into **372 "result families"** covering topics from algebra to theoretical computer science to geometry .

The model that produced these results was an **internal, unreleased system**. OpenAI says it attempted roughly **4,000 problems** and kept the outputs it judged significant .

The average result required computing power equivalent to **about three hours of ChatGPT Pro "thinking"** .

Some of the claims are extraordinary. OpenAI says the model made progress on the **Kakeya conjecture**, advanced algorithms, and even touched on the **Riemann hypothesis** — one of the seven Millennium Prize Problems .

**Only 10 of the 722 manuscripts include reasoning summaries** explaining how the model reached its conclusions. And just **162 of the papers** — about **22%** — come with computer-checked formalizations in Lean, a programming language that mechanically verifies every logical step .

OpenAI itself admits: "some of the unformalized results could have issues" .

In other words, a lot of what was published might be wrong.

---

## The Mathematicians' Dilemma: Excitement vs. Skepticism

The reaction from the mathematical community has been a mix of astonishment, curiosity, and deep concern.

**The Skeptics: "Ask for Receipts"**

Andrew Sutherland, a mathematician at MIT, was blunt: **"Until and unless they release the model and people can replicate their results, I think you should treat any claims about one-shotting problems with a single agent as unverified"** .

He added: **"We should ask for receipts"** .

Daniel Litt, a mathematician at the University of Toronto, was more excited about the potential but shared the concern about process. **"I think that it's great to have new solutions to questions that I and others are interested in,"** he told Fortune. But he worried about the effect on the field if AI is seen as having "solved math" .

**The "Proof Indigestion" Problem**

Terence Tao, one of the most respected mathematicians in the world, coined a phrase that captured the problem perfectly: **"proof indigestion"** .

The idea is simple. Machines can produce mathematical arguments faster than humans can verify, understand, or absorb them. And a proof that no one understands is, in a sense, not really knowledge — it's just a pile of symbols.

Tao wrote on social media: **"Problems are being solved autonomously by AI prompters who have no interest in the broader field itself once their initial target is 'solved', and do not understand the AI output well enough to answer questions on the result, give talks, or otherwise interact with the rest of the field"** .

**The "Slop" Concern**

The Association for Human Mathematics didn't mince words. In a statement criticizing the mass release, it declared: **"Releasing over 700 files at once is not a demonstration of scholarship, but a demonstration of power"** .

The group urged mathematicians to stop collaborating with OpenAI and defend research centered on human understanding.

---

## The Core Problem: Translation and Trust

Here's where the technical issues get thorny.

When AI models solve math problems, they typically produce two things: a **natural language explanation** (the "proof" as written in English) and a **formal verification** in Lean (a machine-checkable code that proves the logic step-by-step).

The assumption is that if the Lean code compiles, the proof is correct. But a new paper by mathematicians at Cambridge and King's College London documented **discrepancies** between OpenAI's natural language proof of a Navier-Stokes-related result and its Lean code .

Specifically, they found that the Lean code proved a **weaker statement** than what the paper claimed. The paper called it "lost in translation" .

**This doesn't necessarily mean the proof is wrong.** But it raises a fundamental question: can we trust AI models to verify their own work? If the model can make mistakes translating its own ideas into code, what else might be slipping through?

Melanie Wood, a Harvard mathematics professor and member of the advisory group, put it simply: **"There is not human understanding of them at the point of release, and now the work begins"** .

---

## Frequently Asked Questions

**Q: What exactly did OpenAI release?**

A: OpenAI published **722 mathematical manuscripts** on GitHub, grouped into **372 result families**. The papers cover topics like algebra, theoretical computer science, geometry, and mathematical physics. They were produced by an internal, unreleased AI model that attempted roughly 4,000 problems .

**Q: Did OpenAI solve the Riemann hypothesis?**

A: **No.** OpenAI did not claim to solve the Riemann hypothesis. The release includes progress "toward" it and other major problems, but no complete solution to a Millennium Prize Problem has been verified .

**Q: Why are mathematicians skeptical?**

A: Three main reasons: (1) **Transparency** — OpenAI hasn't released the model or the prompts used, so results can't be replicated. (2) **Verification** — only about 22% of the papers have computer-checked formalizations in Lean. (3) **Understanding** — even verified proofs may be so opaque that no human understands them, which defeats the purpose of mathematical research .

**Q: What is "proof indigestion"?**

A: The term, coined by Terence Tao, describes the problem of machines producing mathematical arguments faster than humans can verify, understand, or absorb them. It's a problem of volume, not necessarily of quality .

**Q: What is Lean?**

A: Lean is a programming language used to formalize mathematical proofs. When a proof is translated into Lean, a computer can check every logical step mechanically. But passing a Lean check doesn't guarantee the proof is correct — it only shows the formalized version is valid. The formalization itself might contain errors .

**Q: What did the "lost in translation" paper find?**

A: Mathematicians at Cambridge and King's College London found discrepancies between OpenAI's natural language proof of a Navier-Stokes-related result and its Lean code. The Lean code proved a weaker statement than what was claimed in the paper. They say this shows that AI-generated proofs shouldn't be trusted without peer review .

**Q: Are any of these results actually breakthroughs?**

A: Some may be. Daniel Litt of the University of Toronto said he was excited about several results and eager to understand them. Abhishek Saha of Queen Mary University of London called it "a very big day for mathematics," though he noted most of the problems were "exceptional advances within an existing program" rather than game-changing breakthroughs .

**Q: What does the Institute for Advanced Study say?**

A: The Institute for Advanced Study in Princeton, which hosts the Advisory Group on Mathematics and Artificial Intelligence (AGMAI), issued a statement saying it does not endorse OpenAI's practice of testing advanced problems on proprietary models. It said: **"It is now the case that AI can output mathematical arguments in situations without the human who prompted it being able to understand the arguments, verify them, or take responsibility for them"** .

**Q: What should I take away from this?**

A: AI is making genuine progress in mathematics, but the claims require careful verification. The technology is neither good nor bad — the outcomes depend on human choices about oversight and transparency. For now, treat unverified AI-generated proofs with caution .

---

## Conclusion: The Jewels and the Slop

Here's what I keep coming back to when I think about Andrew, the professor with 722 papers on his laptop.

He's not a Luddite. He's not anti-AI. He's a mathematician who cares about understanding — about *why* something is true, not just *that* it is.

OpenAI has produced something remarkable. Buried in those 722 manuscripts are likely genuine discoveries — solutions to problems that have stumped humans for years. Daniel Litt called them "great for mathematics." The Kakeya conjecture progress alone would be a career highlight for any researcher.

But there's also "slop" — unpolished, unverified, possibly wrong. OpenAI itself admits as much.

The real question isn't whether AI can solve math problems. It clearly can. The question is whether the mathematical community can absorb what AI produces without losing the thing that makes mathematics meaningful: **human understanding**.

Terence Tao's "proof indigestion" captures the problem perfectly. We've built a machine that can produce mathematical arguments faster than we can digest them. And a proof no one understands isn't knowledge. It's just symbols.

The jewels are there. The question is whether we can find them — and whether we'll still be able to understand them when we do.

---

## Disclaimer

**This article is for informational and educational purposes only. It does not constitute investment, financial, or academic advice. The author has no position in OpenAI or any related securities. Information presented here is based on publicly available sources and reported statements as of the publication date. The mathematical claims described have not been fully independently verified. Readers should consult official OpenAI communications, academic publications, and qualified mathematicians for the most current and accurate information.**

The Maker of Non-Text AI Model Jev Valued at $7.5B Just Weeks After Launch

 

The Maker of Non-Text AI Model Jev Valued at $7.5B Just Weeks After Launch

**A San Francisco startup you've never heard of just raised nearly a billion dollars. Its AI doesn't chat. It doesn't write. It doesn't generate images. And it might be the most important thing to happen to enterprise AI in years.**

---

## The Model That Refuses to Talk

Let me tell you about a guy named Marcus. He's a software engineer at a mid-sized insurance company in Columbus, Ohio. For the past two years, his team has been trying to automate claims processing using large language models — the same kind of AI that powers ChatGPT.

The problem? The models were too slow. Too expensive. And their outputs were unpredictable. A model might summarize a claim perfectly one day and hallucinate a completely different answer the next.

Then Marcus's team started testing something called **Jev**.

The model didn't write anything. It didn't explain its reasoning. It just returned a number — a probability, a confidence score, a yes-or-no decision. And it did it in milliseconds, at a fraction of the cost.

"It was like switching from a sports car that constantly needs repairs to a bicycle that just works," Marcus told me. "Not as flashy. But it actually gets you where you need to go."

That's the pitch behind TypeSafe AI, the startup behind Jev. And this week, it just convinced some of the smartest investors in Silicon Valley that it's worth **$7.5 billion** .

---

## The Funding Round: $870 Million in Weeks

On October 8, 2026, TypeSafe AI announced it had raised **$870 million** in new funding. The round was led by **Andreessen Horowitz**, with participation from **Sequoia Capital** and existing investor **DCVC** .

The valuation: **$7.5 billion**.

The timeline: **less than a month** after Jev's public launch on September 15 .

To put that in perspective, when TypeSafe emerged from stealth mode earlier this year, it raised a **$40 million seed round** at a valuation of roughly **$200 million** . The company's value has increased **37-fold** in a matter of months.

The funding came after reports that TypeSafe was in talks for at least **$1 billion** in new financing, with some investors offering valuations exceeding **$10 billion** . The final $7.5 billion figure represents a compromise — but it's still one of the fastest valuation jumps in AI startup history.

---

## What Makes Jev Different

To understand why investors are so excited, you have to understand what Jev actually does — and what it doesn't do.

**It Doesn't Generate Text**

Large language models like GPT-6 or Claude are designed to produce human-readable output. You ask a question, they write an answer. You give them a document, they summarize it. You give them a prompt, they generate code.

Jev doesn't do any of that. It doesn't write emails. It doesn't draft reports. It doesn't chat .

**It Generates Decisions**

Instead, Jev produces what TypeSafe calls **"calibrated decisions"** — probabilities, scores, or classifications that software can act on directly .

Here's how it works in practice. Say you're running a customer service platform. You want to automatically route incoming messages to the right department. You describe the categories — billing, technical, general — and set a confidence threshold. Jev reads the message and returns a probability: **92% billing, 6% technical, 2% general**.

Your software can then route the message automatically if the confidence is high enough, or flag it for human review if it isn't .

**It's Built for Automation, Not Conversation**

"We have been super good at human language for four years, but it's not useful for automation because computers speak a different language," TypeSafe co-founder Diogo Almeida told TechCrunch .

That's the core insight. Large language models are optimized for talking to humans. But most AI requests in the future won't come from humans. They'll come from **code** — software programs making automated decisions at scale.

A customer service platform might need to classify 50,000 messages a day. A fraud detection system might need to score 100,000 transactions per hour. An AI agent might need to decide which of 20 possible actions to take, thousands of times per minute.

For those use cases, a model that writes a paragraph of explanation is useless. What you need is a model that returns a number — fast, cheap, and reliably calibrated .

---

## The Numbers Behind the Hype

TypeSafe hasn't published detailed financials, but the company has shared some striking figures.

**Adoption:**

- **One-third of Fortune 500 companies** are already using Jev, according to TypeSafe .

- About **25% of Global 500 companies** are using it, per Almeida's interview with The Wall Street Journal .

- On Vercel's AI Gateway, **nearly 13% of paid teams tried Jev within a day** of its launch — more than **twice as many** as any previous model launch .

**Scale:**

- Jev processes **1 trillion tokens per day**, according to the company .

- The launch video received **40 million views** on X .

**Speed:**

- Jev's response time is **under 700 milliseconds** — roughly **100 times faster** than cutting-edge large language models .

**Team Size:**

- TypeSafe has just **over 20 employees** .

These are remarkable numbers for any startup, let alone one that launched weeks ago. But they also raise questions. The company hasn't published a customer list or usage dashboard to back up its adoption claims. The figures should be read as **reported claims rather than audited facts** .

---

## The Competition Is Already Coming

TypeSafe's success hasn't gone unnoticed. Within weeks of Jev's launch, competitors started rolling out their own decision-focused AI products.

**OpenAI: Decisions API**

On October 6 — just three weeks after Jev's launch — OpenAI released its **Decisions API**, built on its existing GPT-6 Luna model. It's designed for the same kind of classification and routing tasks that Jev handles .

**Databricks: ai_decide**

The data platform company launched **ai_decide**, a feature built into its platform for automated decision-making .

**Cloudflare: Clef**

Cloudflare released **Clef**, an open-source family of models designed for decision tasks .

**Amazon: Strands Decider 2B**

Amazon has a similar product called **Strands Decider 2B** .

Almeida said he expected competition. "It represents the beginning of a new type of AI," he told reporters. "There are still many frontiers to explore. TypeSafe has just taken the first step into the next frontier" .

But the rapid arrival of competitors — especially OpenAI — raises the question of whether Jev's head start is defensible. If OpenAI can build a similar product using its existing models, what's to stop it from dominating the category?

The answer, according to TypeSafe, is **calibration**. Jev was trained from the ground up for decision-making, using a technique the company calls **"reinforcement learning for calibrated decisions"** . Almeida has argued that RLHF — the technique used to train ChatGPT — actually **destroys** a model's calibration ability, making it worse at producing reliable probabilities .

If that's true, Jev has a fundamental architectural advantage. If it's not, the competition could catch up quickly.

---

## Frequently Asked Questions

**Q: What is Jev?**

A: Jev is an AI model developed by TypeSafe AI. Unlike large language models, it doesn't generate text. Instead, it produces probabilities, scores, or classifications — what the company calls "calibrated decisions." It's designed for automation tasks like routing, sorting, and classification .

**Q: Who created Jev?**

A: Jev was created by TypeSafe AI, founded in 2024 by **Diogo Almeida** (former OpenAI researcher), **Sasha Sheng** (former Meta research engineer), and **Erik Gafni** (engineer and entrepreneur) .

**Q: How much did TypeSafe raise?**

A: TypeSafe raised **$870 million** at a **$7.5 billion valuation**. The round was led by **Andreessen Horowitz**, with participation from **Sequoia Capital** and **DCVC** .

**Q: How long after launch did this funding happen?**

A: Jev launched on **September 15, 2026**. The funding was announced on **October 8, 2026** — less than a month later .

**Q: What makes Jev different from ChatGPT?**

A: ChatGPT generates text. Jev generates **decisions**. It returns probabilities or classifications that software can act on directly, without needing to parse a text response. It's faster, cheaper, and designed for automation rather than conversation .

**Q: Is Jev actually being used?**

A: TypeSafe claims **one-third of Fortune 500 companies** are using Jev, and it processes **1 trillion tokens per day**. However, these figures haven't been independently verified, and the company hasn't published a customer list .

**Q: Who is competing with Jev?**

A: OpenAI launched **Decisions API**, Databricks has **ai_decide**, Cloudflare released **Clef**, and Amazon has **Strands Decider 2B**. The category is heating up fast .

**Q: How fast is Jev?**

A: TypeSafe says Jev's response time is **under 700 milliseconds** — roughly **100 times faster** than frontier large language models .

**Q: What is "calibrated decisions"?**

A: It's TypeSafe's term for outputs that express the probability that a result is correct. A Jev output might say "92% confidence this is a billing issue." The calibration matters — if Jev says 90% confidence, it should be right about 90% of the time .

**Q: Is Jev profitable?**

A: Almeida said the company has achieved profitability after accounting for expenses, but didn't disclose revenue data .

**Q: What will TypeSafe do with the money?**

A: The company plans to expand its product lineup, hire more employees, and increase computing resources .

**Q: Should I invest in TypeSafe?**

A: This article is not financial advice. TypeSafe is a private company, so retail investors can't buy shares directly. The valuation is based on private funding rounds and hasn't been tested by public markets. The AI sector is volatile, and early-stage valuations can be highly speculative.

---

## Conclusion: The Quiet Revolution

Here's what I keep coming back to when I think about Marcus, the insurance engineer in Columbus.

For two years, he tried to make large language models work for claims processing. They were too slow, too expensive, and too unpredictable. His team was spending more time cleaning up AI mistakes than saving time.

Then Jev showed up. It didn't chat. It didn't explain. It just made decisions — quickly, cheaply, and reliably.

That's not as exciting as a chatbot that can write poetry. It doesn't make for a viral demo video. But for the thousands of businesses trying to automate routine decisions at scale, it might be exactly what they've been waiting for.

The $7.5 billion valuation says investors believe Jev represents something new — a category of AI designed for machines, not humans. A category that prioritizes speed and reliability over eloquence and creativity.

Whether TypeSafe can defend that valuation against OpenAI and the rest of the competition is an open question. But for now, the company has done something remarkable: it's convinced the smartest money in Silicon Valley that the future of AI isn't about talking. It's about deciding.

---

## Disclaimer

**This article is for informational and educational purposes only. It does not constitute investment, financial, or technology advice. The author has no position in TypeSafe AI, Andreessen Horowitz, Sequoia Capital, or any related securities. Information presented here is based on publicly available sources and reported figures as of the publication date. Adoption claims and usage figures have not been independently verified. The AI sector is highly volatile, and early-stage valuations are inherently speculative. Readers should consult with qualified financial advisors before making any investment decisions.**

The Citroën 2CV Is Back As An Ultra-Affordable City Car

 

The Citroën 2CV Is Back As An Ultra-Affordable City Car

**The new Citroën 2CV is packed with clever retro touches and could become one of Europe's cheapest EVs when it arrives in 2028.**

--

-

## The Car That Mobilized a Nation Is Coming Back

Let me tell you about a man named Jean-Pierre. He's 78 years old, lives in a small village in the Loire Valley, and learned to drive in his father's 2CV. He remembers the way the suspension floated over rutted farm tracks. He remembers the canvas roof rolled back on summer days. He remembers the little air-cooled engine that sounded like a sewing machine and never seemed to quit.

When Citroën announced the 2CV was returning as an electric car, Jean-Pierre didn't believe it. He'd heard rumors before. He'd seen concepts come and go.

Then he saw the photos from the Paris Motor Show. The round headlights. The ribbed hood. The roll-back canvas roof. The tartan seats. The little details that made him laugh — the tortoise-shaped brake pedal, the egg storage in the doors, the removable radio that doubles as a portable speaker.

"It looks like my father's car," he told me. "But it also looks like the future."

He's right. And that's exactly what Citroën is betting on.

---

## The Return of an Icon

On October 10, 2026, Citroën unveiled the new 2CV concept at the Paris Motor Show — the same event where the original 2CV made its debut in 1948 . The French automaker confirmed that a production version will arrive in **2028**, with a target price of **under €15,000** .

That price would make it one of the cheapest electric vehicles on the European market — a direct challenge to the wave of affordable EVs coming from Chinese manufacturers and a bold bet that heritage can sell cars in the electric age.

"The 2CV is back," Citroën CEO Xavier Chardon told investors earlier this year. "Citroën is back. It's back to the future" .

---

## The Design: Retro Without Being Cartoonish

The new 2CV is a masterclass in balancing nostalgia with modern design. It's not a direct copy of the original. It's an interpretation — a modern car that wears its heritage on its sleeve without looking like a museum piece.

**Smaller Than the Original**

Here's something surprising: the new 2CV is actually **smaller** than the car it's based on. It measures just **3.76 meters (147.6 inches)** long, compared to the original's 3.8 meters .

Citroën says this was intentional. The goal was to recapture "the pleasure and agility of smaller cars" while making it easy to park in crowded cities .

The target weight is just **one tonne (about 2,200 pounds)** — remarkably light for an EV. That lightness means Citroën doesn't need a massive battery to achieve useful range, which keeps costs down .

**The Exterior**

The design cues are unmistakable:

- **Round headlights** protruding from the front fenders, now upgraded to LEDs 

- **A ribbed hood** that references the original's distinctive corrugated panels 

- **Steel wheels** — 19-inch rims that are a modern interpretation of the classic design 

- **A black tailgate** that extends down into the bumper, a signature of the earliest 2CVs 

- **A roll-back canvas roof** — the feature that made the original feel open and airy 

The body is finished in a color called **"Infra Red"** — a combination of orange, yellow, and red tones that nods to the classic 2CV Spot .

**The Interior: Simple, Cheerful, and Full of Easter Eggs**

Step inside, and the retro theme continues. But this isn't a car stuck in the past. It's a car that uses simplicity as a design philosophy.

The interior is bright and open, with **wide bench seats** featuring Citroën's Advanced Comfort technology. The passenger seat can slide outward, making it easier to get in and out .

The dashboard is minimalist — a single-spoke steering wheel, a small digital display, and an **8-inch touchscreen** with graphics inspired by comic books . There's no massive screen dominating the cabin. Just the essentials.

And then there are the details that make you smile:

- **A removable storage basket** made of wood and bio-based plastic, replacing the traditional center console 

- **Egg storage** integrated into the doors — a nod to the original 2CV's design brief, which required it to carry a basket of eggs across a rutted field without breaking any 

- **A tortoise-shaped brake pedal** and a **rabbit-shaped accelerator pedal** — playful references to slowing down and speeding up 

- **"Radioën"** — a detachable radio that doubles as a portable Bluetooth speaker 

- **Numbers inscribed on the windscreen** — 48, 90, and 26 — referencing the years the original 2CV started and ended production, and when the new one was revealed 

"The interior definitely doesn't scream 'budget car,' even if it does channel the simplicity of the original," one reviewer noted .

---

## The Tech: What We Know So Far

Citroën hasn't released full technical specifications yet, but the details that have emerged paint a clear picture of what to expect.

**The Platform**

The new 2CV will ride on a **new platform** developed by Stellantis, Citroën's parent company. It will be shared with counterparts from **Fiat and Opel** .

The vehicle will be produced at the **Fiat plant in Pomigliano d'Arco**, near Naples, Italy .

**The Battery and Range**

Industry experts expect the 2CV to use **LFP (lithium iron phosphate) batteries** — a chemistry that's cheaper, more durable, and doesn't rely on expensive materials like cobalt .

The estimated range is around **250 kilometers (about 155 miles)** on the WLTP cycle . That's not a road-trip car. It's a city car — designed for daily commutes, errands, and the kind of driving that most Europeans actually do.

The battery capacity is expected to be between **25 and 30 kWh** .

**The Powertrain**

Details are still scarce, but speculation suggests a single electric motor producing between **60 and 90 horsepower** . That's modest, but it's enough for city driving and short highway trips.

Citroën says the sound of the electric motor at startup is designed to **echo the sound of the original 2CV** . That's a level of attention to detail that suggests Citroën understands what makes this car special.

---

## Frequently Asked Questions

**Q: When will the new Citroën 2CV go on sale?**

A: The production version is scheduled to arrive in **2028** . The concept was unveiled at the Paris Motor Show in October 2026, and Citroën is targeting a development timeline of about 24 months .

**Q: How much will the new 2CV cost?**

A: Citroën is targeting a starting price **under €15,000** (approximately $17,400 or £12,700) before incentives . Some reports suggest the price could be even lower — closer to €10,000 — with government subsidies included .

**Q: Will the new 2CV be available in the United States?**

A: **No.** Stellantis has confirmed that the new electric 2CV will **not** come to the North American market . It's designed for Europe, where small, affordable EVs are in high demand.

**Q: What kind of range will the new 2CV have?**

A: Estimates suggest a range of around **250 kilometers (155 miles)** on the WLTP cycle . This is a city car, not a long-distance cruiser. The lightweight design (targeting one tonne) helps maximize efficiency.

**Q: What battery technology will it use?**

A: Industry experts expect **LFP (lithium iron phosphate) batteries**. These are cheaper and more durable than traditional lithium-ion batteries, and they don't require cobalt — a material with ethical and supply chain concerns .

**Q: Will the production version look exactly like the concept?**

A: Probably not exactly. Concepts typically feature exaggerated elements that don't make it to production. Citroën has said the design "may well evolve" as the car is developed . Some features — like the fairy lights in the roof lining — are unlikely to reach production . But the core retro styling cues should remain.

**Q: What is the "Smart Vision Display"?**

A: It's a bar positioned between the windscreen and the dashboard that shows all the key driving data — speed, range, navigation instructions — directly in the driver's line of sight. It's designed to minimize distraction and keep the dashboard clean .

**Q: Why is it smaller than the original 2CV?**

A: Citroën says the smaller size is intentional, designed to recapture "the pleasure and agility of smaller cars." The compact dimensions make it easier to park and maneuver in crowded European cities .

**Q: What is the significance of the egg storage?**

A: The original 2CV's design brief required it to be able to carry a basket of eggs across a plowed field without breaking any. The egg storage in the new 2CV's doors is a playful nod to that heritage — and a reminder that the car is designed to be practical and capable over rough roads .

**Q: How does it compare to other affordable EVs?**

A: At under €15,000, the 2CV would undercut most competitors. The Renault Twingo starts at around €19,490, and the new Dacia Spring is expected to start under €18,000 . The 2CV's price and retro appeal could make it a unique proposition in the market.

**Q: What is the "Radioën"?**

A: It's a detachable radio styled to look like it comes from a bygone era. It doubles as a portable Bluetooth speaker, so you can take it out of the car and bring the music with you .

---

## Conclusion: A People's Car for the Electric Age

Here's what I keep coming back to when I think about Jean-Pierre and his father's 2CV.

The original 2CV wasn't a luxury car. It wasn't fast. It wasn't beautiful in the conventional sense. It was designed to do one thing: give ordinary people the freedom to move. It was cheap, robust, and endlessly practical. It mobilized a nation.

Citroën is trying to do that again. Not with nostalgia alone, but with a car that's genuinely affordable, genuinely practical, and genuinely charming. A car that costs less than €15,000. A car that's light, efficient, and easy to live with. A car that makes you smile every time you see it.

The electric vehicle revolution has been criticized for catering to the wealthy. Early EVs were expensive, luxurious, and inaccessible to most people. The 2CV is Citroën's answer to that criticism. It's a statement that electric mobility should be for everyone — just like the original 2CV was for everyone.

Of course, there are challenges. The 2CV won't come to America. Its range is limited. Its performance will be modest. And it won't arrive until 2028 — an eternity in the fast-moving EV market.

But for Jean-Pierre, and for millions of Europeans who remember the original, the new 2CV represents something more than transportation. It represents a promise that the future can still be simple, joyful, and accessible.

"The original 2CV gave freedom of mobility to millions of people," Chardon said. "Eighty years later, the new 2CV will democratize electric mobility" .

That's a big promise. But if anyone can keep it, it's the company that built the original people's car.

---

## Disclaimer

**This article is for informational and educational purposes only. It does not constitute investment, financial, or automotive advice. The author has no position in Stellantis (STLA) or any related securities. Information presented here is based on publicly available sources and reported figures as of the publication date. The Citroën 2CV is currently a concept vehicle; production specifications, pricing, and availability are subject to change. Range and performance estimates are projections, not guarantees. Always consult with qualified professionals before making any purchasing decisions.**