Can’t-miss innovations from the bleeding edge of science and techEmail addressSign Up
Last year, a San Francisco-based AI company called Mercor exploded onto the tech scene, just as US workers were buckling under the grip of unemployment.
The online job marketplace connects contractors — often struggling unemployed workers — with companies like OpenAI, the creator of ChatGPT. The gigs have a ghoulish tint, since the basic idea is for former employees to teach AI how to do their old jobs. Adding insult to injury, Mercor is known for treating its workers poorly.
Now, a mini documentary published by More Perfect Unionpulls the curtain back even further on companies like Mercor. In the video, reporter Karen Hao interviews a series of data workers — many of whom wanted to remain anonymous for fear of retaliation — about what it’s like to be employed by what the nonprofit calls “America’s AI sweatshops.”
Perhaps the project’s most eye-catchingly tragic stat comes from 2025 research by the Communication Workers of America, which found that among these tenuous workers training AI systems, a staggering 22 percent said they’d experienced homelessness due to their meager wages.
In the documentary, Hao also cites startling data from a study led by labor researcher Tim Newman. According to his piece, about 86 percent of data workers — those who may be training the AI models you use every day — struggled to pay their bills last year. Nearly one-quarter relied on public assistance programs, such as food stamps and Medicaid.
One interviewee who went by the pseudonym Jen told Hao that she faced an uphill battle in the job market after graduating from an Ivy League school with a PhD more than a year ago. Without any promising career leads, she was forced to move in with her sister and rely on food stamps. Out of desperation, she applied for a gig opportunity from Mercor that offered $55 an hour, far surpassing what she was making as a cashier and substitute teacher.
“I think the role I saw was philosophy intelligence analyst,” Jen told Hao. “I’m looking and I’m like, ‘Well, why wouldn’t I be able to do that?’”
But she soon realized there was reason for skepticism. A mere two weeks after her first project, Mercor entirely pulled the rug out from Jen entirely.
“We all get a message in our group comms where it’s like, ‘Actually, like, this contract is ending,’” Jen explained about her experience.
It turned out there was reason for her to be skeptical — and Mercor’s reported workforce of 30,000 may want to take note.
Photo by Michael Ostuni/Patrick McMullan via Getty Images
In a new substack post, Michael Burry, the hero ofThe Big Short book and movie, declared that the stock market has “jumped the shark,” and posited that “a complete reversal” in the soaring, tech-laden NASDAQ 100 is at hand. Burry noted the resemblance between today’s price action and the waning days of the dot.com craze—adding that it’s feeling like “the last months of the 1999-2000 bubble.” Fellow famed veteran Paul Tudor Jones, in a CNBC interview on May 8, partially echoed Burry’s warning. Jones stated that the current scenario reminds him of 1999, the first year of the infamous furor, noting that if the current momentum keeps rolling, we could be facing “breathtaking kinds of corrections.”
Recommended Video
Wall Street’s analysts and market strategists are at pains to explain what Burry and Jones can’t, namely, why U.S. big caps keep jumping to fresh, all-time records. The economic fundamentals overall look mediocre at best. The current scenario headlines inflation that’s proving both high and extremely sticky, as underlined once again in the April CPI report on May 12 that showed consumer prices advancing a hot 3.7% in the prior 12 months. GDP growth’s ho-hum, the 10-year Treasury yield’s stuck in the elevated mid-4% range, and ultra-tall energy prices keep digging into consumers’ wallets, hiked by a war that keeps dragging on. Not to mention vanishing hopes that the Fed will juice the market via big rate cuts.
In their notes to investors and TV appearances, the bulls increasingly cite the same justification: A surge in corporate earningsthat’s supposed to prove unstoppable due to the super-power of AI. “Absolutely non-stop AI,” Burry fretted after listening to prognosticators in the media endlessly tout the breakthrough as a miraculous cure-all for the sundry negatives. “No one is talking about anything else all day.”
But investors should beware: Torrid earnings-per-share—and they’ve never been this overcooked before—aren’t durable. They come and then inevitably, they go. Profits are subject to huge swings that when they’re unsustainably high, push regular price-to-earnings ratios artificially low, wrongly suggesting stocks are cheap, and when EPS temporarily craters, render PEs abnormally steep, incorrectly portraying the S&P as unusually expensive. Corporate profits now sit at historic peaks as a share of national income, strongly suggesting that in the years ahead, they’ll “revert to the mean” by falling towards the lower, long-term average. That’s always happened when EPS numbers exploded beyond normal bounds in the past. Hence, by swelling the denominator, today’s inflated earnings mask how expensive stocks really are by making PEs appear on the borderline of reasonable.
A highly-respected yardstick erases that illusion. It deploys an averaging system that smooths those temporary spikes and drops in earnings to get a consistent gauge on how richly or lowly-priced stocks really are.
This preferred measure is the renowned cyclically-adjusted price earnings ratio or CAPE developed by Yale professor emeritus and Nobel Prize laureate Robert Shiller. The CAPE’s one of the best forecasters of future returns. When it’s way above the historic norms, you’re likely to get weak returns 5 or 10 years hence; when the measure’s far below average, your chances of prospering in the years ahead are greatly enhanced.
Specifically, the CAPE marshals a 10-year average of inflation-adjusted earnings. That methodology removes the zig-zagging, and calculates a far more accurate PE.
As of May 11, the CAPE had just moved past a dangerous milestone, reaching 40.3. The CAPE—which Shiller posts every month—has only exceeded 40 in its entire 145-year history 21 times, all concentrated in a single continuous period running from January of 1999 to September of 2000. That interlude marked the height of the Dot Com frenzy. Even in the run-up to the Great Depression, the CAPE barely broke 30.
So what kind of returns can you expect buying into the 500 index, or even holding a diversified portfolio of large-valuation U.S. stocks going forward? The record since the end of those 40-plus, Dot Com CAPEs provides a guide: It took twelve years and five months, until February of 2013, for the S&P to regain the super-heady levels of September 2000.
Investors did collect dividends while their cap gains amounted to zero, but all in all, the gains lagged well behind inflation. Investors did much worse than if they’d parked their cash in Treasuries.
The incredible run we’re witnessing may have a simple explanation: Sometimes, markets just go crazy. That argument could be wrong. But it makes just as much sense as the Wall Street hype that paints a gray backdrop as a scene of brilliant sunshine.
The Fortune 500 Innovation Forum will convene Fortune 500 executives, U.S. policy officials, top founders, and thought leaders to help define what’s next for the American economy, Nov. 16-17 in Detroit. Apply here.
The big picture: Established datingapps and new startups are using AI to overcome the swipe fatigue that’s forced the online dating industry to innovate. Through AI-assisted conversation starters, in-app assistants and AI-powered chemistry testing, the tech has many uses in the business of love.
Bumble CEO Whitney Wolfe Herd teased the app’s AI assistant “Bee” coming later this year in an interview with Axios.
Zoom in: New York-based Amata coordinates some 2,000 first dates a month. Users who agree to the AI matchmaker’s pairing purchase a $20 “date token,” and the app plans the details.
To discourage ghosting, the app builds in consequences:If you cancel two dates in a row, you’re temporarily blocked from matching.
“It’s really focused on intentional dating,” Amata spokeswoman Mandy Menaker says.
Another approach: Carly Malatskey founded SoCal-based AI matchmaker Joey AI after noticing the dating startups she encountered through her venture capital work lacked nuance. “People are choosing a life partner … as mindlessly as scrolling on TikTok,” she says.
With Joey, there’s no swiping. There’s not even an app. It starts with a phone call between an interested single and their AI matchmaker.
I gave Joey a ring. In a mellow Australian accent, the AI asked me my name, job and basic dating preferences, then went deeper: How important is politics in my relationships? What time did I wake up today? How often have I talked with my family this week?
After that initial call, users are verified and photos are shared,with Joey connecting new hopeful romantics via text. (I opted out of getting matched — a journalist engaged to her high school sweetheart likely isn’t the target audience.)
“Joey starts as a matchmaker and then can grow into this wingman,” Malatskey says, with users reaching out to Joey for advice — and pep talks — as dates proceed.
For San Francisco-based Known, there’s no in-app chatting between users, no profiles and no swiping.
Users talk to an AI matchmaker and pay $15 to secure their real-life hang, which also helps prevent no-shows
.The goal, co-founder and CEO Celeste Amadon says, is to feel like you’re being introduced by a friend who “understands you really, really well, but knows everybody in your city instead of a couple hundred people.
“Case in point: Marie Lansley, a 36-year-old San Franciscan, tried out Known to find her Prince Charming. She was struck by the matchmaker’s emotional intelligence, and appreciated not having to build a profile.Her first match wasn’t love at first bot, but she’s not ruling out letting AI find her true love. “I’m not 100% sure it can right now, but maybe it can help me sift through the volume so that I can then go out and meet that person,” she said.
The bottom line: “Chemistry will always be analog,” Lansley says.
Photo illustration: Sarah Grillo/Axios. Photo: Andrew Harnik/Getty Images
President Trump flew to Beijing today under some of the darkest economic clouds of his political career, departing a country reeling from the cost of everyday life, Axios’ Zachary Basu writes.
Why it matters: The bottom is falling out on Trump’s economic credibility — the central promise of his return to power. The inflation crisis that doomed his predecessor suggests he may not recover.
A new CNN pollfound that 70% of Americans disapprove of Trump’s handling of the economy — a benchmark that never crossed 50% in his first term, even during the pandemic.
77% of Americans, including a majority of Republicans, say Trump’s policies have driven up the cost of living in their own community.
For now, Trump appears unconcerned, convinced that renewed inflation is temporary and that gas prices will plummet once he ends the Iran war.
Asked before departing for China whether Americans’ financial struggles were motivating his push for a deal with Iran, Trump replied: “Not even a little bit.”
“The only thing that matters when I’m talking about Iran is they can’t have a nuclear weapon,” he added. “I don’t think about Americans’ financial situation.”Data: The Silver Bulletin. Chart: Noah Bressner/Axios
The big picture: The affordability crisis that fueled Trump’s return to power has become a five-alarm threat to his presidency — even as GDP growth, largely thanks to the AI boom, remains strong on paper.
1. Prices are surging: Inflation spiked to 3.8% in April as the Iran war pushed the national average price of gas above $4.50 a gallon.
Economists fear the energy shock is beginning to ripple through the broader economy, pushing up the cost of groceries, airfare and electricity.
2. Paychecks are shrinking: Yesterday’s inflation report showed that prices are outpacing wages for the first time in three years, erasing gains in real purchasing power.
American households have absorbed a nearly 30% rise in consumer prices since the pandemic — a cumulative wound that has never fully healed, Axios’ Courtenay Brown reports.
3. Debt is mounting: Americans are increasingly leaning on credit cards and loans to absorb rising costs, with consumer borrowing posting its biggest monthly jump in March since late 2022.
The personal savings rate fell to 3.6% in March, its lowest level since 2022, as lower-income households burn through savings to cover essentials.
4. Confidence is collapsing: Consumer sentiment has cratered to record lows as Americans grow pessimistic about the economy and their own financial futures.
A newYouGov/Economist poll found 59% say the economy is getting worse, while just 15% say it’s improving. More than two-thirds of Americans say the country feels “out of control.”
5. Main Street is souring: The National Federation of Independent Business says optimism around future business conditions and expansion plans has fallen to its lowest level since before Trump’s reelection.
Small businesses are often treated as an economic early-warning system because they’re especially vulnerable to rising fuel costs, tighter credit and weakening consumer demand.Share this story.
Bloomberg defense tech reporter Katrina Manson, author of a new book on Project Maven, discusses what the IDF’s use of AI in Gaza reveals about the gap between US and Israeli standards for civilian harm. Manson draws on investigative reporting from 972 Magazine, as well as her own conversations with US military officials who analyzed the IDF‘s approach.
That reporting found very little time between AI-surfaced intelligence and actual strikes. US officials told Manson that Israel had not broken the laws of armed conflict, but was “leaning in towards autonomous processes in a way that the US had never leaned anywhere near as close.”
What this reveals is a stark difference in how two allied militaries weigh civilian harm. Manson says the IDF was prepared to accept higher casualty numbers per mid-ranking target than US military tradition allows. That gap is a policy question as much as a technology one; as Manson and Drew Kukor, the former chief of Project Maven, both acknowledge, AI and policy are “absolutely interlinked”.
Harvard releases information on 1,613 enslaved individuals
Public database advances research on University’s ties to slavery, bolsters effort to help descendants recover family histories
Jacob Sweet
Harvard Staff Writer
May 12, 2026 6 min read
Harvard has published a database identifying 1,613 people who were enslaved by University leaders, faculty, or staff or who labored on campus as enslaved individuals between 1636 and 1865.
The publicly accessible Harvard Slavery Remembrance Program (HSRP) database is an update on the University’s research, and a result of a recommendation included in the 2022 Report of the Presidential Committee on Harvard & the Legacy of Slavery. The report initially identified more than 70 individuals. The new HSRP database includes the names, locations, and documented dates of enslaved people — as well as the names and positions of the Harvard affiliates who enslaved them. The research behind the database is being led by American Ancestors, the nation’s oldest genealogical nonprofit and the research partner of the Harvard & the Legacy of Slavery (H&LS) Initiative.
“Harvard and our partners have approached this work thoughtfully, seriously, and with respect for those individuals we are able to identify and the family histories we can help recover,” said Sara Bleich, vice provost for special projects at Harvard and leader of the H&LS initiative. “To expand our research from just over 70 individuals to now 1,613 has taken genealogical expertise on the part of countless researchers. And, while our work is by no means done, this is a big step forward.”
The database is the product of rigorous genealogical and archival research. While genealogical research often begins with a living person and traces backward, for enslaved individuals, “We do the opposite: start in the past and move to the present,” said Lindsay Fulton, chief research officer at American Ancestors. “We are basically doubling the research — because you have to research both the enslavers and the people they enslaved.”
To find the descendants of people who were enslaved by Harvard leaders, faculty, or staff, researchers first built out a list of who held those positions in the years between 1636 and 1865. The University didn’t have a centralized staff registry until much more recently, which meant researchers had to comb through handwritten notes from University meetings, stewards’ books, faculty records, legislative charters, and a variety of other sources to recreate Harvard’s roster from the ground up. Through this work, researchers have verified approximately 3,000 members of leadership, faculty, or staff, creating a framework where none had previously existed.
“My hope is that, over time, unflinching self-examination will ripple outward, that Harvard will be a leader not only in scholarship but in demonstrating institutional honesty and humility in confronting the complexities of our institutional past.”Henry Louis Gates Jr.
“In researching people who were enslaved by Harvard affiliates, we first needed to understand the structure of the University, the different positions people held, and how these changed over time,” Fulton said. “For example, members of the Board of Overseers were often appointed because they held positions within the colonial government or because they were church ministers. But the criteria for who was an overseer changed over time.”
From there, researchers searched for documentation that indicated which individuals enslaved people. This information could lead to uncovering the names, or in some instances where names were not apparent, indications of those they enslaved. The new database identifies 259 members of Harvard’s leadership, faculty, or staff prior to the end of the Civil War who enslaved individuals. American Ancestors’ research into these 259 and other Harvard leaders, faculty, and staff is ongoing and expected to grow significantly.
Performing simultaneous genealogical work for the Harvard leaders, faculty, or staff who enslaved individuals as well as those they enslaved requires diligence and attention to detail. For each of the former, researchers examined a specific set of documents, including probate records, land and property deeds, and marriage records, among many more.
Identifying enslaved individuals, who were considered property under colonial and pre-Civil War law, can be even more complex. These individuals are often mentioned only in passing in estate disputes that can stretch several hundred pages. In some cases, their names shift over time.
While the database represents a major expansion from the approximately 70 names included in the 2022 report, the growth does not come as a surprise. The presidential committee had anticipated that the list would widen considerably as the H&LS Initiative implemented Recommendation 4 from the report. The H&LS Initiative was established in 2022 to implement the seven recommendations the committee detailed in the report.
Harvard and American Ancestors acknowledged that the database is far from finished; researchers will continue to identify more individuals enslaved by University leadership, faculty, or staff — and trace the descendants, living and deceased, of those they enslaved. While the work to recover and reconstruct family histories and family trees will take time, so far researchers have identified about 600 living descendants. The H&LS Initiative will continue sharing new findings with the public at key milestone moments, helping support a wider effort of institutions exploring their ties to slavery. The University will contribute this research to the 10 Million Names project, a collaborative initiative led by American Ancestors that is dedicated to recovering the names of the estimated 10 million men, women, and children of African descent who were enslaved in pre- and post-colonial America.
“My hope is that, over time, unflinching self-examination will ripple outward, that Harvard will be a leader not only in scholarship but in demonstrating institutional honesty and humility in confronting the complexities of our institutional past,” said Alphonse Fletcher University Professor Henry Louis Gates Jr., who is also a member of the initiative’s Advisory Council. “Every chapter in history, every family tree, and every institution, has its share of shadows and surprises. The journey isn’t always neat and easy, but it’s a crucial part of self-knowledge — an experience both necessary and transformative.”
To explore the HSRP database, learn more about the research methodology, and review resources for pursuing genealogical research, visit the Harvard & the Legacy of Slavery website.
May 10, 2026 NicIreland is often presented as one of Europe’s great economic miracles: soaring GDP, huge multinational tax receipts, and some of the world’s richest headline numbers. But behind the statistics, ordinary people face absurd rents, unaffordable real estate, overcrowded hospitals, weak infrastructure, and a state dangerously dependent on a handful of global giants. In this video, we look at Ireland’s distorted wealth story, the multinational tax model behind it, and why so much money has failed to translate into a better life.
To discourage ghosting, the app builds in consequences:If you cancel two dates in a row, you’re temporarily blocked from matching.
Users talk to an AI matchmaker and pay $15 to secure their real-life hang, which also helps prevent no-shows
The goal, co-founder and CEO Celeste Amadon says, is to feel like you’re being introduced by a friend who “understands you really, really well, but knows everybody in your city instead of a couple hundred people.
The bottom line: “Chemistry will always be analog,” Lansley says.
Data: 
The big picture: The affordability crisis that fueled Trump’s return to power has become a five-alarm threat to his presidency — even as 
Menu
Search
canisgallicus 