A reminder for new readers. That Was The Week collects the best writing on critical issues in tech, startups, and venture capital. I select the articles because they are of interest. The selections often include things I disagree with. The articles are only snippets. Click on the headline to go to the original. I express my point of view in the editorial and the weekly video.
Content this week from: @geneteare, @lucyrbrewster, @denniskneale of
Last week’s editorial was titled Gone. This week’s is titled Shrinking. By some measures, that is an improvement. But last week's object of study was dying business models, DVDs, Magazines, and Buzzfeed. This week’s focus is Venture Capital. And it certainly has not gone and is not going.
But it is shrinking.
Crunchbase News carries an overview of Seed, Series A, B, and C round shrinkage, both my number of rounds and the average and median raises at those rounds. But the more poignant is a second sister piece on Fintech. In that piece, Gené Teare summarizes where we are in the correction that began in private markets last year. The emphasis is mine.
Now let’s look at companies that have raised funding — anywhere from seed to Series C — by year but have yet to raise funding in subsequent years. These are the companies in danger of running out of runway if they don’t get additional capital.
Around 900 seed fintechs globally that raised at least $1 million in funding have not raised funding since 2021, an analysis of Crunchbase data shows. Another 1,400 seed-stage companies raised at least $1 million in a single seed funding in 2022 and have not raised again in 2023.
Of the Series A through Series C funded companies that raised funding in 2021, about 1,000 — around 65% — have not raised since 2021. A further 1,100 companies from Series A to Series C raised funding in 2022, but not again.
For companies funded at seed through Series C in 2019 and 2020, by contrast, roughly a third did not raise funding in subsequent years. Those companies have had more years in which to raise follow-on funding, but were also funded in a climate where median fundings and the absolute number of companies funded were lower.
Connect this to the points concluded in her first article and a picture emerges:
While it is clear we are not going back to funding levels from a decade ago, the question is whether round sizes will shrink back to pre-pandemic levels.
For now, the only stage we analyze here that’s dropped below 2020 levels is Series C funding. But even Series C is still above the 2019 median and average round size.
What we can say is that the reset is only two to three quarters in and likely has not yet hit bottom. Each distinct funding stage is reacting to the cuts in the stage after it. If late-stage funding continues to contract due to the closure of the IPO markets, then startups at earlier stages face an uncertain future.
This correlates to Redpoint Ventures’ analysis from a couple of weeks ago.
On this chart, your eyes naturally go to the downward curves in 2000, 2008, and 2022. But the real story is the flat bars that follow for several years. The normal recovery after a deep correction is over ten years.
Now, there are many important drivers in 2023 that did not exist in prior cycles. Over 4 billion internet users, the emergence of AI and its value creation potential, CRISPR and other BioTech innovations driving human health, quantum computing, new sustainable materials due to chemical and biological science progress, and possibly new cheap energy. All of these will provide ample sources of growth and value. Venture Capital will recover, possibly quickly.
We cannot even rule out that rational investing behavior will lead to a new bubble, especially in AI. But the power to attract capital at scale will belong to companies doing their Series A in 2023 and after.
Those who raised seed and Series A in 2020 and 2021 will, as Gené says, need to consider big valuation changes to qualify for 2023 B, C, and D rounds. That is a lot of companies. It will lead to more failures than has been normal. Or, to put it another way, the slowdown on Series B for 2020, 21, and 22 Series A companies is not just a slowdown. Without action, it is a signal that they may never do a Series B. That would imply there are two ecosystems now in startupland. One that raised a Series A before 2023 and one that has raised since. The second will be part of the new normal.
Like all projections, this one will be wrong in many cases, so if you are one of those founders, don’t panic. Simply ask what needs to be done and do it.
Many things that were true in 2022 and before are no longer true. Here is a partial list:
There will be over 100 unicorns every year
The time between funding rounds will often be less than 12 months for good companies
You need an 18 month runway
Assumptions about valuations at each round and amount of dilution
Assumptions about IPO timing
Multiples of ARR needed to achieve a valuation
Growth investors will fight to get into the best B,C, and D rounds
Opportunity funds are a good idea
Funds will raise every 2-3 years
There are many more of these former truths that are no longer true. Knowing them is key to navigating the next phase. In that phase, capital is sparse, the quality bar is high, and past investors may not follow on. The quality of execution will be more important than ever, as well as the scale of the target market.
It is likely that the class of 2023 seed and Series A rounds will produce some of the biggest companies yet. Knowing which they are and engaging with them will produce very well-performing funds. More in this week’s video on Saturday.
The size of the typical seed funding round peaked in 2022, Crunchbase data shows, but has since dipped.
If you go back about 10 years to 2014, the median and average seed funding for a U.S.-based startup was below $1 million.
Since 2014, the typical seed deal has increased in size and peaked in 2022 at a median of $2.5 million and an average of $3.7 million.
In contrast to the overall venture funding pullback in 2022, seed funding was higher in the first half of 2022 compared to 2021 and showed a decline year over year starting in the fourth quarter.
Series A
The median and average Series A deal size also peaked in 2022 before dropping more recently, Crunchbase data shows.
The growth of the seed ecosystem impacted Series A fundings as it expanded the number of years a startup could build in advance of a Series A funding.
Median and average Series A fundings in 2014 were $5 million and $7.7 million, respectively, for U.S.-based startups.
The typical Series A deal size peaked in 2022 at $14 million (median) and $19.1 million (average). It has since come down to $12 million and $18.7 million, respectively — not a big drop, one could say. (But, keep in mind, we expect those amounts for the most recent quarter to trend down further as fundings continue to be added to the Crunchbase database after the end of the quarter. We find that deals added after the close of a quarter tend to be smaller.)
Series B
Series B fundings peaked sooner than seed and Series A — all the way back in 2021, Crunchbase data shows. This indicates a sharper pullback in funding in 2022 from Series B onward.
In 2014, Series B deals tracked at a median of $11.7 million and an average of $16.3 million.
That ramped up substantially over the following decade to peak in 2021 at a median of $32 million and an average of $46 million.
That’s dipped again somewhat to $28 million and $40 million, respectively, in 2023.
Series C
Similar to Series B funding, Series C peaked in 2021. But the jump in funding happened in 2020, a year earlier than prior funding stages.
Series C fundings show the greatest decline for all stages analyzed here. In 2014, the median Series C funding was $18 million and the average $26.4 million. That peaked in 2021 at $60 million and $82 million, respectively.
In Q1 2023, a median Series C round for a U.S.-based startup was $42 million and the average $59 million. The gap has also narrowed slightly, indicating fewer larger rounds at Series C.
Never going back again
While it is clear we are not going back to funding levels from a decade ago, the question is whether round sizes will shrink back to pre-pandemic levels.
For now, the only stage we analyze here that’s dropped below 2020 levels is Series C funding. But even Series C is still above the 2019 median and average round size.
What we can say is that the reset is only two to three quarters in and likely has not yet hit bottom. Each distinct funding stage is reacting to the cuts in the stage after it. If late-stage funding continues to contract due to the closure of the IPO markets, then startups at earlier stages face an uncertain future.
Few sectors illustrate the massive runup in venture funding that occurred in 2021 as well as financial services and the fintech industry. In that year, billion-dollar venture fundings went to neobanks, wealth management providers, buy now, pay later startups, cryptocurrency exchanges and insurance brokers.
Some 20% of the total $681 billion in global venture funding in 2021 went to the fintech sector alone.
But only a small handful of these companies — including Robinhood and Nubank — successfully went public. The majority of the highly funded fintechs from 2021 are still private and face cost cutting in a tougher sales environment, all while waiting for the public markets to turn favorable again.
While much of the focus has been on late-stage companies, which are closer to the public-market turmoil, there is also a huge backlog of seed and early-stage fintech startups that were funded during the heady days of 2021, but haven’t raised new funding since.
That could mean a massive funding shortfall for the fintech sector over the next four to eight quarters, an analysis of Crunchbase data shows.
Let’s dive in.
Fintech’s stellar rise — and fall
First, let’s chart the growth and decline in recent years.
Venture funding to financial services companies hovered around $10 billion per quarter on a global basis in 2019 and 2020, then shot up to $30 billion and then $40 billion in two back-to-back quarters in 2021, Crunchbase data shows. Fintech funding then settled for three quarters around $34 billion before slipping to $25 billion in Q2 2022 and then dwindling all the way back to around $10 billion in the fourth quarter.
Fintech funding dipped more than overall global venture funding in 2022, falling 40% year over year compared to overall global venture funding, which slipped 35%.
Q1 2023 was back up above $10 billion, but that was in large part thanks to a single massive deal: Stripe’s $6.5 billion raise.
Early stage grew too
The surge in fintech funding in 2021 was most noticeable at late-stage. But the increase also impacted companies at the early stages — and it wasn’t only large outlier deals. Median Series A fundings grew by 47% in 2021 and Series B by 85%.
Back in the good old days for Silicon Valley in 2021, unicorns, private companies worth over a billion, became more abundant and bigger than ever before. But with the IPO market largely shuttered, M&A was expected to take off in early 2023. But that hasn’t happened. With high inflation, tech companies cutting costs, and a crackdown by antitrust regulators, M&A activity across stages and sectors has nosedived—meaning yet another exit ramp is blocked off for many startups. According to PitchBook-NVCA Monitor, acquisitions of venture-backed startups saw their lowest quarterly level in a decade.
And those worth the most are having serious trouble finding buyers. Currently, there are 704 active unicorns with an aggregate post-money valuation of about $2.4 trillion, according to PitchBook. “Unicorns are likely to face significant challenges when sourcing potential acquirers due to limited buyer interest, and the recent decline in M&A activity coincides with the current economic downturn—only $39.6 billion in U.S. acquisition value has occurred since the beginning of 2022, which makes it the least active year since 2015,” explained PitchBook venture capital analyst Vincent Harrison in today’s research note about M&A among unicorns.
Companies are largely choosing to hold off on buying even potentially profitable startups due to high inflation and pressure to cut costs. What are they turning to instead? “So far, share buybacks have been favored over M&A, with buyback announcements hitting a record high of $1.2 trillion in 2022,” explained Harrison.
Another reason ambitious tech companies are wary of purchasing a unicorn is the antitrust push from President Biden and the federal government to stop companies from amassing huge swaths of market share. Meta is currently being sued by the FTC for antitrust violations related to its acquisitions of WhatsApp and Instagram. “This crackdown is likely to continue, affecting cash-starved unicorns seeking acquisition by large corporations,” explained Harrison. “So much so that many would-be acquirers may choose not to pursue deals altogether, further limiting exit opportunities available to unicorns and reducing their bargaining power and valuation, he added.
The bottom line? Once high-flying unicorns could start hitting the market for cheap. “The combination of limited exit options and the need for private capital may drive some unicorns to accept discounted acquisition offers as a means to either survive and/or generate some sort of return for themselves and investors,” said Harrison.
Thanks to innovative new features and a trimmed-down staff, ads and money are coming back in droves.
WSJ Opinion by Dennis Kneale
April 19, 2023 5:52 pm ET
Elon Musk in San Francisco, Jan. 24. PHOTO: BENJAMIN FANJOY/ASSOCIATED PRESS
Elon Musk is building Twitter into something better, bigger, safer and, above all, freer. Yet media headlines claim the company is “dying.”
The doom is overwrought. The eccentric billionaire is creating one of the most powerful media platforms the world has ever seen. This was underscored last week when he schooled an officious BBC reporter who claimed—without evidence—that the amount of hate speech is up on the social-media site. The CEO’s quick-witted lesson on free expression has since made the rounds on the platform, racking up well over 20 million views. That’s impressive reach—as was the company’s performance during the World Cup, when it hosted 147 billion impressions in about four weeks.
Mr. Musk is now accelerating plans to turn the chat room into the “everything app,” a marketplace he calls X.com, where users can buy and sell stocks, open bank accounts, shop for retail goods, pay bills and more. Last week he renamed Twitter’s holding company X Corp. and moved its incorporation from Delaware to Nevada. He also formed a new Nevada-based company, X.AI, and hired a former scientist from Google’s DeepMind laboratory to lead the venture. And Mr. Musk signed a deal with an Israeli online stock trader called eToro, which boasts more than 32 million users across Europe, Asia and the U.S.
In December, Twitter added pricing data for $Cashtags: a function that enables users to type a dollar sign before a stock symbol—say, $TSLA—to see the latest tweets about the company. So far this year, CNBC reported, Twitter users have logged some 420 million searches using the feature. Now the platform will allow Twitter users to click cashtags and jump to eToro to buy or sell stock.
Meantime, Mr. Musk is introducing more capabilities to the app: “We’re releasing features faster than in Twitter’s history—at the same time as having contained the costs and reduced the cost structure by a factor of three, maybe four,” he told the tech podcast “All-In” in December. Twitter’s “views” function is one such, which offers constant updates of how many people saw and clicked on each tweet. Mr. Musk noted that this requires Twitter’s servers to crunch some three million calculations per second—every second.
Another new feature, Community Notes, lets selected contributors rate and add factual context to posts. The site states that if enough contributors from “different points of view rate [a] note as helpful,” it will be publicly affixed to a tweet. No longer can journalistic “fact-checkers” get away with the shoddy work typical of the genre. One recent example: When the Washington Post’s Glenn Kessler tweeted what he described as the “incendiary claim” that George Soros “funds” New York prosecutor Alvin Bragg, he was instantly met with popular—and accurate—pushback.
All the while, Mr. Musk has cut a workforce of 8,000—with a median compensation of $232,626 a year—down to about 1,500, drawing the envy of his counterparts in Silicon Valley. The media responded by claiming that hate speech and vile images would soar on the site because of the cuts. Yet all indications suggest that cracking down on this content is a priority of Mr. Musk’s.
The name of the podcast I co-host with Kara Swisher is “Pivot.” I don’t like the name, but I’ve had my hands on the wheel for so long at my own companies, I’m down with sitting in the backseat and occasionally asking, “Are we there yet?” Besides, Kara does most of the work and has a better feel for pods than me. But that’s not what this post is about.
A “pivot” is a strategic change in business model, direction, or target market. Think Netflix’s shift from DVDs to streaming, Adobe’s move to subscription, or Amazon’s launch of AWS. Sounds easy, but real transitions require a staggering investment and a leap of faith that make shareholders queasy. And, most of the time, they don’t work. Meta’s stock doubling in the last six months is a function of the market’s belief that The Zuck is waking up from his Big Gulp, Venti Grande Ayahuasca hallucination re a $20 billion-per-year investment in the metaverse. Meta’s earnings this week revealed Reality Labs (its opium den posing as a business unit) saw revenue decline to $350 million while losing $4 billion in the last 3 months. This means the birth control known as Oculus is pacing to lose the combined profits of Honda, BMW, and General Motors in ’23.
These bets are dwarfed by the greatest redirect in economic history: The Gulf States’ attempt to pivot from oil-based economies to something more sustainable.
Party’s Over
For the past century, the Gulf States have run on oil — and still do. State-owned oil giant Saudi Aramco is now more valuable than the next 10 largest energy companies combined, and last year it booked $161 billion in profits — likely the largest net income figure ever recorded.
Only, there’s a catch. The well is running dry. Regardless of the cadence of the move to renewables, the battery running the Gulf EV goes dead some time this century. Data re the oil remaining under the sand are closely guarded state secrets for the Gulf nations. But Bahrain is expected to run out within the decade, Oman in two decades, and the Kingdom by the end of the century, possibly sooner.
The plan to redirect this wealth into something else is unprecedented: building a next-generation civilization from scratch. Scratch, plus a few trillion dollars.
Its Own Moon
The scale and boldness of this bet is peerless. Let’s start with a 105-mile-long glass-domed mega-city in the desert to house 9 million people with no cars, staffed by robots, and powered entirely by wind and solar. Oh, and it will have a ski resort. And its own moon.
This sci-fi mega-city is the centerpiece of Saudi Arabia’s Neom project, budgeted at $500 billion. Keep in mind, that’s the budget — and 9 out of 10 mega-projects go over budget. Saudi Arabia is also building the Diriyah Gate, a $20 billion property development that will add 20,000 homes to the historic district of Diriyah, and the Red Sea Project, which will build 1,000 homes and 50 hotels across 22 small islands. Meanwhile, Qatar is building its own “city of the future” fit with a 90,000-capacity sports stadium, a dedicated entertainment district (“Entertainment City”), and the country’s first six-star hotel. No ski resort, though.
Bait
The secret sauce in the Gulf State pivot isn’t the oil money itself, however. That’s the bait. The prize is other people’s money. Specifically, rich people’s money. The plan, distilled, is to become the global headquarters for the mega-wealthy. This strategy increasingly makes sense as wealthy people continue to weaponize even democratic governments whose policies crowd more of the spoils into the top .01%.
The best way to attract the rich is to give them what they want. Which in 2023 means three things: luxury hospitality, world-class entertainment, and low taxes. The Gulf has gone all-in on all three.
The Saudis have launched their own ultra-luxury hotel brand, the Boutique Group, and Qatar owns The Plaza Hotel in New York and The Savoy in London. (Pro-tip: These weren’t property or even hotel acquisitions but brand acquisitions.) The United Arab Emirates built the iconic Burj Al Arab and has its own luxury hospitality university. Ever fly Emirates, Etihad, or Qatar airlines? (Think in-air shower spa.)
Grimes has welcomed musicians to create new songs with her voice using Artificial Intelligence, saying she would split 50% of royalties on any successful AI-generated track that included her voice.
The Canadian singer, whose real name is Claire Boucher, tweeted that it was the “same deal as I would with any artist I collab[orate] with. Feel free to use my voice without penalty,” she tweeted.
She said she was interested in being a “guinea pig” and she thought “it’s cool to be fused with a machine and I like the idea of open sourcing all art and killing copyright”.
The music industry is currently entering unparalleled territory as it tries to keep up with the implications of a spate of songs created by training AI to generate artists’ voices.
Last week, Universal Music successfully petitioned TikTok, YouTube and Spotify to remove a track titled Heart On My Sleeve, which used AI vocals generated from their artists Drake and the Weeknd.
It was just one of several recently released tracks that featured AI-generated vocals based on Drake, who does not seem to be as enthused as Grimes. The rapper recently wrote: “This is the final straw AI,” on an Instagram story, referring to a version of Ice Spice’s song Munch that was released with a fake verse by him.
In a statement, the label said “the training of generative AI using our artists’ music” was “a violation of copyright law”. However, Universal’s position has not been tested in court, and it remains a legal grey area whether art that is created by a human, but which contains AI elements, can be copyrighted.
In October, the Recording Industry Association of America (RIAA) warned that AI companies were violating copyrights en masse by using music to train their machines.
A snapshot of what happened this past year in AI research, education, policy, hiring, and more.
Apr 3, 2023
Shana Lynch
The 2023 AI Index is out, covering the world of artificial intelligence from technical performance achievements, ethics advances, education and policy trends to economic impact, R&D, and the hiring and jobs scene.
The AI Index is an independent initiative at the Stanford Institute for Human-Centered Artificial Intelligence (HAI), led by the AI Index Steering Committee, an interdisciplinary group of experts from across academia and industry. It tracks, collates, distills, and visualizes data relating to artificial intelligence, enabling decision-makers to take meaningful action to advance AI responsibly and ethically with humans in mind.
TL;DR? Here, learn about the state of AI in 14 charts.
1: LLMs Scale Up
Large language models keep scaling in size and expense. GPT-2, released in 2019 and considered the first large language model, had 1.5 billion parameters and cost an estimated $50,000 to train. Just three years later, PaLM launched with 540 billion parameters and cost an estimated $8 million. It’s not just PaLM: Across the board, large language and multimodal models are becoming larger and pricier. (And since these are estimates, we've qualified them as mid, high, or low: mid where the estimate is thought to be a mid-level estimate, high where it is thought to be an overestimate, and low where it is thought to be an underestimate.)
If you only have a few minutes to spare, here’s what investors, operators, and founders should know about AI’s history.
Architecting logic. In the summer of 1956, Herbert Simon and Allen Newell showed off their remarkable program “Logic Theorist” to a collection of enlightened peers, only to be met with indifference. In the years since, Logic Theorist – which was capable of proving complex mathematical theorems – has been recognized as the first functional AI program. Its use of structured, deductive logic was an example of “symbolic AI,” an approach that dominated in the following decades.
The rules of the world. AI’s most glorious failure may be a project named “Cyc.” In 1984, Douglas Lenat began his attempt to create a program with an understanding of our world. The then-Stanford professor sought to develop this context by inputting millions of rules and assertions the AI could use to reason – including basics like “all plants are trees.” Lenat believed insufficient knowledge represented a huge barrier for AIs that was best solved with thoughtful, manual intervention. Despite decades of development and hundreds of millions in investment, Cyc has struggled to deliver meaningful results.
Embodied AI. In the 1980s, a new school of practitioners emerged to challenge AI’s dogma. This group, led by Australian academic Rodney Brooks, argued that real intelligence wouldn’t come from assiduously designed logical frameworks but by allowing machines to take in sensory input and learn from their environment. This “embodied” approach to AI ushered in practical robots, albeit with narrow applications.
Emulating the brain. Geoff Hinton is regularly cited as the “godfather” of modern AI. The University of Toronto professor earned that honorific through his steadfast belief that powerful intelligence would be achieved by modeling the patterns of the brain. Hinton’s contributions to “neural networks” – a structure directly based on our gray matter – paved the way for modern AI systems to flourish.
Learning by doing. How do you create an AI capable of beating the world’s greatest Go player? In the mid-2010s, Cambridge startup DeepMind showcased the potential of a radically new learning technique called “reinforcement learning.” Rather than learning how to play chess or Go through a set of strict rules, DeepMind’s engines developed by playing the game and receiving positive or negative feedback on their actions. This methodology has driven advancements far beyond the Go board.
In their classic work, The Lessons of History, husband and wife Will and Ariel Durant analyzed the story of human civilization. Among their pithy and profound observations is this meditation: “The present is the past rolled up for action, and the past is the present unrolled for understanding.”
No aspect of our present moment feels more ready to act upon the fabric of our lives quite as radically as artificial intelligence. The technology is accelerating at a pace that is hard to comprehend. A year ago, crowd-sourced predictions estimated artificial general intelligence would arrive in 2045; today, it’s pegged at 2031. In less than a decade, we could find ourselves competing and collaborating with an intelligence superior to us in practically every way. Though some in the industry perceive it as scaremongering, it is little wonder that a swathe of academia has called for an industry-wide “pause” in developing the most powerful AI models.
To understand how we have reached this juncture and where AI may take us in the coming years, we need to unroll our present and look at the past. In today’s piece, we’ll seek to understand the history of AI through seven experiments. In doing so, we’ll discuss the innovations and failures, false starts and breakthroughs that have defined this wild effort to create discarnate intelligence.
Before we begin, a few caveats are worth mentioning. First, we use the term “experiments” loosely. For our purposes, an academic paper, novel program, or whirling robot fit this definition. Second, this history assumes little to no prior knowledge of AI. As a result, technical explanations are sketched, not finely wrought; there are no equations here. Thirdly, and most importantly, this is a limited chronicle, by design. All history is a distillation, and this piece is especially so. Great moments of genius and ambition exist beyond our choices.
One final note: This piece is the first installment in a mini-series on the foundations and state of modern AI that we’ll add to in the weeks and months to come. We plan to cover the field’s origins and technologies, and explore its most powerful companies and executives. If there are others you think would enjoy this journey, please share this piece with them.
Last week I wrote a piece noting how my wife Michelle’s Google usage was down by nearly two thirds, thanks to her discovery of ChatGPT. I noted that Michelle isn’t exactly an early adopter – but that’s not entirely true. Michelle is more of a harbinger – if an early tech product “fits” her, she’ll adopt it early and often – and it’s usually a winner once it goes mainstream. The early Tivo DVRs come to mind – and they remain a better product than anything that’s come since in the television world (another example of how entrenched business models kill innovation).
But few early versions of any new product get to “Michelle market fit” on first attempt. For it to happen with an AI chatbot – well before I developed the habit – is rarer still. I mean, I’m supposed to be the early adopter around here!
So once I noticed Michelle was hooked, I asked her how she wrangles ChatGPT. As I noted in my last post, the majority of her usage focuses on information-intensive projects that tend to get messy when attempted with Google. For example, Michelle’s managing a real estate project with a complex set of inputs. The property requires extensive renovations but must conform to certain community regulations and standards. With the summer rental season approaching, she needs to make scores of decisions on everything from paint colors to septic system contractors. Prior to ChatGPT, Michelle would have started her searches inside Google, but quickly been frustrated by “content cruft” – reams of crappy results and, in her words, “way too many ads.” For commercial searches about household appliances, construction projects, landscaping ideas, and the like, Google’s index favors the kind of content created by “advertising mills” – cheap, low calorie stuff that crowds out the kind of trusted advice Michelle is seeking. She knows she’s being steered into choices that create profit for Google’s ecosystem.
Faced with those kinds of results, Michelle is more likely to call a friend who has already done what she’s looking to do, get one simple answer (even if it’s not exactly right), and go with it. And this is the first insight that springs to mind when I talk to Michelle about ChatGPT. For her, the service has become the equivalent of a “first phone call” for a competent point of view, minus the cruft. It’s as if OpenAI has culled most of the crap from the content mills that have larded up Google, and delivers only the good stuff.
Continuing along those lines, Michelle tells me she’s found a hack of sorts – she’ll ask ChatGPT for that competent first answer, then ask it to name top design blogs that focus on her particular query. Then she’ll dig into those sites, and more often than not, she’ll find confirmation of ChatGPT’s advice, or sometimes even better ideas. This is how, for example, she figured out which shade of paint would work for the property she’s working on.
Last spring, engineers in Barcelona packed up the sperm-injecting robot they’d designed and sent it by DHL to New York City. They followed it to a clinic there, called New Hope Fertility Center, where they put the instrument back together, assembling a microscope, a mechanized needle, a tiny petri dish, and a laptop.
Then one of the engineers, with no real experience in fertility medicine, used a Sony PlayStation 5 controller to position a robotic needle. Eyeing a human egg through a camera, it then moved forward on its own, penetrating the egg and dropping off a single sperm cell. Altogether, the robot was used to fertilize more than a dozen eggs.
The result of the procedures, say the researchers, were healthy embryos—and now two baby girls, who they claim are the first people born after fertilization by a “robot.”
“I was calm. In that exact moment, I thought, ‘It’s just one more experiment,’” says Eduard Alba, the student mechanical engineer who commanded the sperm-injecting device.
The startup company that developed the robot, Overture Life, says its device is an initial step toward automating in vitro fertilization, or IVF, and potentially making the procedure less expensive and far more common than it is today.
Right now, IVF labs are staffed by trained embryologists who earn upwards of $125,000 a year to delicately handle sperm and eggs using ultra-thin hollow needles under a microscope.
But some startups say the entire process could be carried out automatically, or nearly so. Overture, for instance, has filed a patent application describing a “biochip” for an IVF lab in miniature, complete with hidden reservoirs containing growth fluids, and tiny channels for sperm to wiggle through.
Scientists might soon be able to create eggs and sperm from skin and blood cells. What will that mean?
“Think of a box where sperm and eggs go in, and an embryo comes out five days later,” says Santiago Munné, the prize-winning geneticist who is chief innovation officer at the Spanish company. He believes that if IVF could be carried out inside a desktop instrument, patients might never need to visit a specialized clinic, where a single attempt at getting pregnant can cost $20,000 in the US. Instead, he says, a patient’s eggs might be fed directly into an automated fertility system at a gynecologist’s office. “It has to be cheaper. And if any doctor could do it, it would be,” says Munné.
MIT Technology Review identified a half-dozen startups with similar aims, with names like AutoIVF, IVF 2.0, Conceivable Life Sciences, and Fertilis. Some have roots in university laboratories specializing in miniaturized lab-on-a-chip technology.
So far, Overture has raised the most: about $37 million from investors including Khosla Ventures and Susan Wojcicki, the former CEO of YouTube.
"We're becoming more efficient so we can build better products faster and put ourselves in a stronger position to deliver our long term vision," said CEO Mark Zuckerberg.
Meta Platforms (META) - Get Free Report posted better-than-expected first quarter earnings late Wednesday, while forecasting solid near-term revenues, thanks in part to resilient ad spending and surprise boost in active users on its most-important platform.
The Facebook parent said profits for the three months ending in March were pegged at $2.20 per share, down 19.1% from the same period last year but firmly ahead of the Street consensus forecast of $2.03 per share.
Group revenues, Meta said, rose 2.65% to $28.65 billion, nearly all of it -- $28.1 billion -- coming from the new 'Family of Apps' division the company created last year, just ahead of analysts' estimates of a $27.41 billion tally. The year-on-year revenue gain was the first advance since 2021.
Ad impressions rose 26%, Meta said, although the average price per ad was down 17%. Monthly active users across Meta's 'Family of Apps' was tabbed at 2.99 million, up 2% from last year, while daily active users were up 4% from last year at 2.03 billion, just ahead of the Street's 2.01 billion estimate.
Looking into the current quarter, Meta said it sees revenues in the region of $29.5 billion to $32 billion, a range that fall under the Street forecast of $32.3 billion.
First Republic (FRC) - Get Free Report shares turned sharply lower Friday, and were halted from trading on the New York Stock Exchange, amid reports that federal regulators are looking for rescue solution that would involve sending the lender into receivership.
CNBC reported Friday that the Federal Deposit Insurance Corporation (FDIC) is exploring a way in which First Republic can be sold to another U.S. bank, but added that a sale would involve taking the lender into receivership and having its assets and franchise were transferred to the eventual buyer.
The report noted that officials were still hopeful of a so-called 'open markets' solution, which would not involve a First Republic wind-down, but those prospects appeared to be fading into the weekend.
Purchasing the assets in a receivership situation gives buyers more leeway in terms of picking and choosing valuable assets, including loans and mortgages, while an 'open market' solution likely forces the buyer to pay above-market rates for the hundreds of millions in Treasury bonds sitting on First Republic's balance sheet.
Reuters has also reported that Federal Reserve and Treasury officials are working with the FDIC in an attempt to broker a financial lifeline for the bank, which saw deposits fall by more than $100 billion over the first three months of the year.
Humane, the startup founded by ex-Apple employees Imran Chaudhri and Bethany Bongiorno, has given a first live demo of its new device: a wearable gadget with a projected display and AI-powered features intended to act as a personal assistant.
Chaudhri, who serves as Humane’s chairman and president, demoed the device onstage during a TED Talk, a recording of which has been acquired by Inverseandothers ahead of its expected public release on April 22nd.
“It’s a new kind of wearable device and platform that’s built entirely from the ground up for artificial intelligence,” Chaudhri says in comments transcribed by Inverse. “And it’s completely standalone. You don’t need a smartphone or any other device to pair with it.”
Thanks to the presentation, we now have at least some idea of what the device might be able to do and how it might go about doing it without a traditional touchscreen interface. During the presentation, Chaudhri wears the device in his breast pocket, tapping it in lieu of a wake word and then issuing voice commands like you would with an Amazon Echo smart speaker. Axios notes that the device also supports gesture commands. ..
Humane co-founder Imran Chaudhri appearing on screen wearing a prototype wearable computer in his jacket pocket. Photo: Ina Fried/Axios
Ex-Apple employee Imran Chaudhri gave TED attendees on Thursday an early glimpse of the AI-powered wearable that his startup, Humane, has been developing.
Why it matters: The screenless device, which does not require a nearby cell phone to work, uses a combination of voice and gestures for input and can display information by projecting it onto nearby objects.
Details: In his TED talk, Chaudhri showed the wearable, which sat in his jacket pocket, translating his own voice into French.
He also answered a phone call from his wife with the call information appearing as a green image projected onto his hand.
"This is good AI in action," he said, promising more details would be released in the coming months.
The secretive startup has raised $230 million, including $100 million announced in March, with investors including Kindred Ventures, SK Networks, LG Technology Ventures, Microsoft, Volvo Cars Tech Fund, Tiger Global, Qualcomm Ventures and OpenAI CEO and co-founder Sam Altman, per TechCrunch.
