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 my editorial and the weekly video.
Oh boy, what a day. 9 March 2023 will never be forgotten at Silicon Valley Bank.
The after-market share price fell to $77.78, a 71% decline from the previous closing price.
This week’s curated stories below explain what happened. The short version is that the bank took some steps to strengthen its balance sheet and, in the process, baked in some losses from having placed customer funds into bond investments. The losses were about $1.8 billion. A large loss for the bank.
Adding fuel to the fire, SVB announced a capital raise from multiple sources, signaling a 30% dilution for existing investors.
The immediate impact on March 8th was a 20% or so decline in SVBs share price. By market opening, it had increased to 30% and is now at 71%. Friday will test whether this represents a bottom.
Meanwhile, in Palo Alto, the atmosphere is confused. Many venture funds called for calm. Others asked their companies to withdraw funds. As I write, it is impossible to withdraw funds from SVB due to system failures, so there are lines at the physical branches.
One thing, however, is certain. The bank leadership, aware of impending corrections, made some loss-making decisions that have stunned investors and resulted in panic share selling.
The fate of SVB is tied to the fate of many investors and entrepreneurs in Silicon Valley. It is also tied to the reputation of the Valley as a beacon of risk-taking and innovation.
Whether the bank survives through the efforts of its friends, is sold to a “white knight” buyer, or fails altogether is open to question.
SignalRank banks with SVB, and we have many friends there. We wish them the best as this plays out.
Enjoy this weeks reading, even though it may be challenging for those with a stake in the outcome.
Essays of the Week
March 9, 2023 12:02 PM PST
Silicon Valley Bank CEO Greg Becker on Thursday told top venture capitalists in Silicon Valley to “stay calm” amid concerns around a capital crunch that wiped nearly $10 billion off the bank’s market valuation.
On a call, Becker said that “calls started coming and started panic.” He added that the bank has “ample liquidity to support our clients with one exception: If everyone is telling each other SVB is in trouble, that would be a challenge.”
“I would ask everyone to stay calm and to support us just like we supported you during the challenging times,” he said.
Image Credits: Frank Herholdt / Getty Images
It may go down in the history books about Silicon Valley: the time that its most prominent bank, a bank founded nearly 40 years earlier, inflicted such grievous injury on itself that it had to be rescued by another bank or else risk going down in flames in a single day.
We don’t yet know who that “white knight” will be, but you can bet a lot of conversations are happening right now about who will step in and acquire Silicon Valley Bank, an institution whose shares are down roughly more than 80% in after-hours trading from where they were at the beginning of yesterday. And why? Not because the bank is falling apart at the seams. Instead, because it utterly flubbed some important messaging at the very worst time imaginable.
This, friends, is what is called an own goal.
If you’re just catching up, here’s what happened: Silicon Valley Bank lost $1.8 billion in the sale of U.S. treasuries and mortgage-backed securities that it had invested in, owing to rising interest rates. The bank is also contending with shrinking customer deposits, given that its customer base of largely startups has far less money right now to park at a financial institution……
A Great Thread by Genevieve Roch-Decter
It sold off a $21 billion bond portfolio for a huge loss to shore up liquidity.
The market is worried and the stock is down 54% today.
Here’s what you should know 🧵
Another Good Thread, by Turner Novak
Image Credits: dem10/ Getty Images
Some of Silicon Valley Bank’s customers are struggling to transfer funds out of their bank accounts, numerous sources tell TechCrunch.
The seeming wave of attempted withdrawals comes after SVB announced yesterday that it lost $1.8 billion in the sale of U.S. treasuries and mortgage-backed securities that it had invested in, owing to rising interest rates. The bank also said that it was raising more capital, and investing into higher-yield products. Concern ensued, leading the share price to tank more than 50% at time of publication.
Dozens of VCs are advising their portfolio companies to pull their assets from the bank, sources say, while others are pushing for founders to at least diversify where they hold their capital. Others, meanwhile, warn that the panic is coming too early — perhaps from earlier news this week that Silvergate, another bank, is shutting down. SVB as a result is clearly experiencing deposit volatility from a subset of its users.
March 9, 2023
Silicon Valley Bank — the preeminent bank for tech startups and venture debt in the valley — saw its shares plummet Thursday after announcing it would sell billions of dollars in stock to shore up its balance sheet and cut its outlook for the year.
The bank’s shares fell more than 60% Thursday, closing around $106 — well off its 52-week high of $597. That led to a broader decline among bank stocks, with the Nasdaq Bank Index falling 8%.
The significant drop led to deep concern around the venture world — SVB has relationships with more than 50% of all venture-backed companies in the U.S. and countless VC firms — around the bank’s liquidity and balance sheet strength.
Index of banks posts biggest drop since pandemic roiled markets nearly three years ago
Greg Becker, chief executive of SVB, tried to reassure venture capitalists and other clients in the tech sector.PHOTO: LAUREN JUSTICE/BLOOMBERG NEWS
Updated March 9, 2023 6:55 pm ET
Investors dumped shares of SVB Financial Group and a swath of U.S. banks after the tech-focused lender said it lost nearly $2 billion selling assets following a larger-than-expected decline in deposits.
The four biggest U.S. banks lost $52 billion in market value Thursday. The KBW Nasdaq Bank Index notched its biggest decline since the pandemic roiled the markets nearly three years ago. Shares of SVB, the parent of Silicon Valley Bank, fell more than 60% after it disclosed the loss and sought to raise $2.25 billion in fresh capital by selling new shares.
Banks big and small posted steep declines. PacWest Bancorp fell 25%, and First Republic Bank lost 17%. Charles Schwab Corp. fell 13%, while U.S. Bancorp lost 7%. America’s biggest bank, JPMorgan Chase & Co., fell 5.4%.
Thursday’s rout is another consequence of the Federal Reserve’s aggressive campaign to control inflation. Rising interest rates have caused the value of existing bonds with lower payouts to fall in value. Banks own a lot of those bonds, including Treasurys, and are now sitting on giant unrealized losses.
Large declines in value aren’t necessarily a problem for banks unless they are forced to sell the assets to cover deposit withdrawals. Most banks aren’t doing so, even though their customers are starting to move their deposits into higher-yielding alternatives. Yet a few banks have run into trouble this week, sparking fears that other banks could be forced to take losses to raise cash.
The sudden popularity of generative AI has re-generated a popular pre-pandemic preoccupation: How many jobs will AI destroy?
Some prediction experts predicted a decade ago that almost half of U.S. jobs could be replaced by AI by 2023 (!) or, at most, by 2033, mainly impacting low-skill jobs (e.g., no more truck drivers because we will have self-driving trucks). Other crystal-ball observers argued that in contrast to previous waves of automation, we are entering a new era in which the most affected will be highly-skilled knowledge workers.
The tight labor market of recent years has suppressed somewhat these dire predictions. The widespread excitement about generative AI, however, is bringing back the anxiety about jobs, especially the creative kind of jobs.
According to the Harris Poll, “most workers are wary of generative AI,” and 50% don’t trust the technology. The Atlantic tells us confidently: “No technology in modern memory has caused mass job loss among highly educated workers. Will generative AI be an exception?… While it is difficult to predict the exact extent of this trend, it is clear that AI will have a significant impact on the job market for college-educated workers.”
White-collar workers may soon face the AI disruption everyone’s been panicking about. But the news may be better than you think.
ANYONE WHO HAS spent a few minutes playing with ChatGPT will understand the worries and hopes such technology generates when it comes to white-collar work. The chatbot is able to answer all manner of queries—from coding problems to legal conundrums to historical questions—with remarkable eloquence.
Assuming companies can overcome the problematic way these models tend to “hallucinate” incorrect information, it isn’t hard to imagine they might step in for customer support agents, legal clerks, or history tutors. Such expectations are fueled by studies and media reports claiming that ChatGPT can get a passing grade on some legal, medical, and business exams. With companies like Microsoft, Slack, and Salesforce adding ChatGPT or similar AI tools to their products, we are likely to see the impact on office life soon enough.
A couple of research papers posted online this week suggest that ChatGPT and similar chatbots may be very disruptive—but not necessarily in the ways you expect.
First up, Edward Felton and colleagues at Princeton University try to identify the professions most likely to be affected by ChatGPT. They used a benchmark called the AI Occupational Exposure, which maps occupational tasks to the capabilities of various AI programs, to see which jobs are most vulnerable to chatbots with strong language skills.
The results suggest big changes may be ahead for those in some professions, including telemarketers, history teachers, and sociologists, while people with more physical occupations, such as brickmasons, dancers, and textile workers, may not need to worry about ChatGPT showing up at their place of work.
But a second study suggests that people in language-centric jobs are not necessarily destined for replacement. Shakked Noy and Whitney Zhang, graduate students at MIT, looked at what happens when you put ChatGPT in the hands of office workers. They asked 444 college-educated professionals to complete a series of simple office tasks, including writing press releases and short reports, drafting emails, and creating analysis plans. Half of them got to use ChatGPT.
The study found that people with access to the chatbot were able to complete the assigned tasks in 17 minutes, compared to an average 27 minutes for those without the bot, and that the quality of their work improved significantly.
As machine learning becomes core to every product, engineering teams will restructure.
In the past, the core engineering team & the data science/machine learning teams worked separately. The core engineering team ships the product & focuses on reliability.
For most companies, the data science team analyzed data & wrote machine learning models to support the business functions : sales, marketing, customer support. These teams operated downstream of the data warehouse. But they are movin’ on up to work on products in front of the data warehouse.
ChatGPT has catalyzed an expectation in users’ minds that all products should converse with users & should be rather intelligent.
To accomplish this vision, companies will merge the data science team into core engineering.
Data scientists’ expertise will be required to optimize the end-user experience : determining the right data sets to use to train models, understanding the best prompts to elicit germane answers (prompt-engineering), & model monitoring. They will need to be part of core engineering to achieve those goals.
The forces behind this change have been brewing for some time. Data science teams’ increased budget faces pressure from the office of the CFO to justify the investment. LLM-features should contribute directly to revenue via upsell & market share, quieting questions.
Also, the challenges of pushing models into production have frustrated data science teams. In this new world, data scientists will commit directly to production code.
Organizational change will require behavioral change. Machine-learning enablers – startups that sell ML toolkits – have shifted sales tactics to seize the day. Their ideal customer profile is no longer data science teams but core engineering teams.
March 9, 2023 6:00 AM PST ·
Elon Musk fanned a growing culture war in artificial intelligence by confirming last week that he plans to develop an “anti-woke” alternative to OpenAI’s ChatGPT, as The Information first reported.
Musk and other critics have a point, said Greg Brockman, OpenAI’s co-founder and president. In an interview, he said the startup did not move quickly enough to give users greater ability to customize the behavior of the chatbot, which has been criticized for inaccuracies and has faced claims that its responses reflect a left-leaning political bias. “We made a mistake,” Brockman said. “Our goal is not to have an AI that is biased in any particular direction.” Still, he said there are some lines AIs should never cross.
• OpenAI built ChatGPT as a step toward artificial general intelligence
• President Greg Brockman favored making ChatGPT public to find mistakes and blind spots
• “There should be some hard bounds that AIs will never cross”
Brockman, who was chief technology officer of Stripe before co-founding OpenAI in 2015, has been a key technological force inside the company since its creation. In the interview, he discussed why he pushed to develop a chatbot as a step toward creating software systems that can think and learn on their own, as humans do, a concept known as artificial general intelligence, AGI. And he revealed that OpenAI had difficulty developing a viable product, considering and discarding several ideas along the way. “We had a technology in search of a problem,” he said.
This interview has been edited for length and clarity.
The Information: Why did you develop an AI assistant?
The $650 million startup isn’t a newsletter platform – it’s a multi-media network and cultural powerhouse.
Mario Gabriele, Mar 5
If you only have a few minutes to spare, here’s what investors, operators, and founders should know about Substack.
More than a newsletter platform. Though Substack has come to define the newsletter movement, its scope, and ambitions extend far beyond it. Chris Best, Hamish McKenzie, and Jairaj Sethi have built more than a simple way to send emails and collect payments, constructing a powerful creative network.
A product transformed. The market downturn complicated Substack’s plans to raise a Series C. While that necessitated cost-cutting, the chillier climate seems to have had some benefits. In 2022, Substack shipped extraordinary improvements to its product, none more important than its recommendations feature, which has driven huge growth for top writers.
Real revenue and upside. Substack’s Series B valued it at $650 million. It’s unlikely the firm would command that price today – though it doesn’t mean the business isn’t in good shape. Extrapolating from Substack’s available data, the company is likely earning north of $20 million in revenue with a low burn. Given Substack’s rapid growth and revenue opportunities, we should expect good days ahead.
The wheels must spin. To date, Substack has delivered for its top writers, driving more revenue than the 10% take rate it clips. Recommendations were critical in that equation last year. The company must ensure it keeps existing flywheels spinning and adds new ones. Publications will take their subscriber lists elsewhere if Substack fails to justify its fees.
Creating cultural power. It isn’t easy to value something as intangible as cultural power, but Substack has it. Important news, analysis, and narratives are shared on the platform, and some of the world’s most influential people use it to connect with their audiences. That influence is difficult to compete with and could help make the company a true category creator.
Moses Beach needed a faster horse. In 1846, the publisher of The Sun, then one of New York City’s largest dailies, sought to deliver timely news about the Mexican-American War. It was, after all, the story of the day, dominating public attention. “Has the Mexican War terminated yet, and how? Are we beaten?” a teenage Emily Dickson wrote to her brother, expressing the nation’s inquisitorial mood.
Though Beach understood the appetite for updates from the battlefront and recognized the value strong coverage could bring to The Sun, he faced a conundrum: how could he acquire the best information quickly without incurring enormous costs? Though telegraphs had been invented, the nearest machine was in Richmond, Virginia, thousands of miles from the frontline. Could The Sun afford to run a relay of horses for days, weeks, or even years?
Ever the innovator – Beach had dabbled in gunpowder-powered balloons and steamships before turning to the publishing business – he devised a clever solution. Rather than footing the cost himself, he invited four other New York newspapers to go in with him. Together, they would split the costs; together, they would benefit from faster information.
Beach’s plan worked. This chain of colts and stagecoaches propelled The Sun to new heights and illustrated the power and leverage media companies could produce by acting as a network: sharing infrastructure costs and collaborating rather than strictly competing.
Though designed to solve a temporary problem, that structure would have enduring value. In the years following the Mexican-American War, Beach’s consortium became The Associated Press. Today, the “AP” is one of the most trusted, wide-spanning media organizations on earth, both a “news agency” – distributing content to other publishers – and a destination in its own right. Every day, it produces 1,000 stories, reaches 250 countries, and is read by half the world’s population.
Glance at Substack and its parallels with a nearly two-hundred-year-old non-profit are not obvious. Beneath the surface, though, there is a common signature uniting the organizations, a shared business allele. Fundamentally, Substack exists to defray infrastructure costs for publishers, make it easier for them to build their businesses, and drive them to new heights. Rather than a monolithic entity, it’s a network in which publishers collaborate, helping one another at least as much as they vie for attention. And, if Substack has its way, it will evolve into a premier destination of its own, sought out by hundreds of millions of readers – perhaps touching billions in total.
News of the Week
March 8, 2023
Global monthly funding fell to $18 billion in February 2023. Not since February 2020 has global funding dipped below $20 billion in a single month.
Late-stage funding fell the most, by 73%, while early-stage funding was down 52% year over year, based on an analysis of Crunchbase data.
This drop in late-stage funding takes place in a climate in which venture funds are closing on billion-dollar funds. This past month, San Francisco-based Bain Capital closed on $1.9 billion in funds, its largest to date. And data-driven investor SignalFire, also based in San Francisco, raised $900 million across its seed, breakout and opportunity funds, larger than prior funds.
More than a year after tech stocks plunged, have we hit the market bottom yet?
In a recent interview with TechCrunch, Social Capital founder and “SPAC king” Chamath Palihapitiya said his firm has cut back on late-stage funding due to lack of opportunity. He assesses it will take three years to price late-stage fundings accurately as investors resist downgrades in portfolio valuations.
Money deployed at Series B, Series C and Series D in the past two years will struggle to raise follow-on funding. In this funding environment, we have reported on an increase in round extensions as companies struggle to raise the next funding at an increased valuation.
March 7, 2023
When technology and biotech valuations began to tumble last year, U.S. Series B funding held up at first, offering one of the brighter areas in a more bearish investment environment.
That’s no longer the case.
In recent months, Series B funding to U.S. startups has fallen sharply. So far in 2023, investment is on track to come in at the lowest quarterly level in more than three years.
The continuing slowdown this year comes after dealmaking had already been contracting for a couple quarters. Per Crunchbase data, Series B investment in the second half of 2022 was down over 60% from the year-ago period.
For a sense of how far things have fallen, we chart out Series B deal counts and funding for the past nine quarters below.
Why Series B?
So why single out Series B for scrutiny?
We periodically look at Series B trends because this stage functions as a sort of proxy for the broader venture funding ecosystem: It’s typically the point in a startup’s lifecycle when it has proven technology or market traction and is on the cusp of major scaling.
So far in 2023, Crunchbase data shows investors are backing fewer big rounds. There were just six Series B deals of $100 million or more to date this year, an average of fewer than one per week. In 2022, by comparison, there were 90 such rounds, or almost two per week.
March 6, 2023
Five companies joined The Crunchbase Unicorn Board in February — the fourth month in a row for new unicorns to number in the single digits.
Among these new unicorns is the first company from Egypt to join the board: Cairo-based fintech MNT-Halan, which provides digital wallets as well as buy now, pay later offerings and business loans. The service has 1.3 million monthly active users with more than 5 million customers in Egypt. With this strong foothold, MNT-Halan plans to expand to surrounding countries in the region.
A further three companies joined from the U.S. and one company from China.
A year ago, in February 2022, 56 companies joined the Unicorn Board, before new unicorn numbers progressively dropped throughout the year.
Funding to current unicorn companies was also down in February, tracking at $2.9 billion raised by 27 companies, including the new unicorns.
Current unicorns total 1,439 companies on a global basis that are collectively valued at $4.9 trillion with $852 billion raised in funding, per the Crunchbase Unicorn Board.
March 9, 2023 2:13 PM PST ·
If there’s one indication of how difficult the fundraising environment has become, it’s that limited partners—the wealthy individuals and endowments that invest in VC funds—are trying to sell their stakes in those funds. In other words, not only are LPs not interested in your next venture fund, they want out of your last one.
“There’s a rare LP out there that wants to add venture today,” said Hans Swildens, the founder and CEO of Industry Ventures, an investment firm that buys secondary stakes in companies and other venture funds. Rather, “there’s a lot of people trying to offload everything.
“Almost every single company and every single fund has shareholders and LPs that are looking for liquidity,” he said.
There are a few reasons LPs want out. Some are plagued by the denominator effect because the value of their public stocks have declined so much that their portfolio now leans too heavily into private assets. They need to offload VC stakes to rebalance the holdings. And others have realized that exiting now, before VC fund values have fallen too much, offers them the best chance of recouping most of their investment.
Swildens spoke to me by phone this week from Tiburon, Calif., across the bay from San Francisco, as he watched sperm whales leaping in the water. He says the net asset value of venture funds came down about 20% between the first quarter and the third quarter in 2022, on average. This is because companies—Stripe, Instacart and Reddit among the high-profile examples—were overvalued during the boom and have since been marked down by investors. That’s similar to the decline in 2009, but it pales in comparison to the 90% drop in fund values in 2001.
For those like Swildens who’ve made a career out of buying scraps, it’s a great time to be in business. Last year, Swildens said, he put hundreds of millions into secondaries, and he hopes to triple that number this year.
The CEO of social-media giant Meta has sworn by AI, popularized by the chatbot ChatGPT.
LUC OLINGA, MAR 2, 2023 10:27 AM EST
There will be no press release, no big announcement, as he would have to acknowledge that he was wrong.
But make no mistake: Mark Zuckerberg just buried the metaverse. The metaverse is dead.
The metaverse was supposed to be the Next Big Thing for the social-media tycoon, who in 2021 went so far as to rename his empire — created from Facebook, Instagram and WhatsApp — as Meta Platforms.
Simply put, the metaverse is an immersive virtual world in which we are supposed to interact with each other using specialized glasses and virtual-reality headsets.
It was the future of technology, according to Zuckerberg, whom Tesla CEO Elon Musk dubbed “Zuck the Fourteenth” in an apparent nod to the French king Louis the XIV, famous for his hubris and excess.
Meta Creates a Top-Level AI Team
For those who doubted the company’s devotion to the idea, Meta has invested billions of dollars in this massively hyped project — to the chagrin of company shareholders.
In 2021 and 2022, Reality Labs, the division housing metaverse projects, recorded a cumulative loss of nearly $24 billion, including $13.7 billion just last year.
The losses will ease significantly in coming months because the metaverse is over.
Zuckerberg has just held the funeral by turning to the next big shiny thing, namely artificial intelligence…
PUBLISHED WED, MAR 8 202311:18 PM EST
Customers hold shopping bags outside the Shein Tokyo showroom in Tokyo on Nov. 13, 2022. Reuters reports the fast fashion retailer is targeting a U.S. IPO in the second half of 2023.
Noriko Hayashi | Bloomberg | Getty Images
Chinese online fashion retailer Shein is set to raise around $2 billion in a new funding round this month and is aiming for a U.S. listing in the second half of this year, three people with knowledge of its plans told Reuters.
The United Arab Emirates’ sovereign wealth fund Mubadala is a major investor in this round as are existing investors, private equity firm General Atlantic (GA) and venture capital group Sequoia Capital China, said two of the people and a separate person with knowledge of the matter.
Tiger Global Management became a new investor, said the first two people.
Shein cut its valuation to $64 billion in this fundraising, down by a third from a funding round a year ago, according to six sources with knowledge of the matter.
The company last month held initial talks with several investment banks to pick lead bookrunners for the U.S. IPO, said two of the sources with direct knowledge of the plans.
In another move to monitor the rise of AI within China, the Chinese government has ordered large technology companies in the country to refrain from providing services related to ChatGPT. According to NikkeiAsia, this directive is part of a broader effort to tighten control over online content and prevent the spread of what the government deems as harmful information. State media also blasts ChatGPT as a spreader of U.S. government misinformation.
Regulators have told tech giants, Tencent Holdings and Ant Group, an Alibaba Group Holding affiliate, to not offer access to ChatGPT services on their platforms. This goes for direct services and third parties as reported by sources speaking to NikkeiAsia. This is no surprise as the popular chatbot is currently unavailable in China without a virtual private network.
Because of this, Tencent suspended many third-party services recently, even if said services were not connected to ChatGPT or any similar programs. But it wasn’t only the ban on the use of ChatGPT. Regulators also told tech companies that they must report any ChatGPT-like programs to Beijing before launch. This is unsurprising as the Chinese government regularly bans foreign websites and apps for servicing within the country given its strict control of online content.
Startup of the Week
Amazon sold close to $40bn of advertising last year – bigger than Prime, bigger than the entire global newspaper industry and probably more profitable than AWS. But is this really advertising, rent, or something else? And what does that mean for Google?
I first started paying attention to Amazon’s ad business a couple of years ago, when a line called ‘other’ in the back of the accounts started to grow really fast. In 2021 Amazon started splitting it out, and last year Amazon had $38bn of ad revenue, more than any traditional media company and bigger than the entire (now much reduced) global newspaper industry.
The simplest way to look at this is as a very profitable new business. $38bn is a small percentage of Amazon’s overall $502bn 2022 revenue, but it probably had well over 50% operating margins, which would mean it brought in as much operating profit as AWS – $20-25bn, and without anything like as much capex. (Amazon’s overall operating income, meanwhile, was only $12bn.)
More interesting, perhaps, is the fact that $38bn is also more than reported revenue for Prime, which also points to the problem with trying to calculate a stand-alone profitability for this. Prime has directly attributable revenue and costs, but it’s also an important marketing tool, and it drives increased purchasing: the real revenue from Prime is a lot more than subscription fee. In the same way, calculating an actual P&L for Amazon Ads would be pretty artificial (this is also why Apple doesn’t generate a P&L for the app store, even internally). …continued