Istio is the last major component of the Kubernetes ecosystem to sit outside of the CNCF. At one point, Google seemed unsure whether it would cede control of the service-mesh project, but things have changed.

Google’s announcement comes almost exactly two years after Google Cloud CEO Thomas Kurian told Protocol that the technology giant planned to donate Istio to a foundation.

Google plans to announce Monday that it will submit the Istio open-source service mesh project and trademark to the Cloud Native Computing Foundation (CNCF).

Google executives told Protocol that the Istio Steering Committee will ask for Istio to be admitted as an incubating project within the CNCF, the vendor-neutral home to several open-source enterprise tech projects including Kubernetes, the popular open-source container orchestration system originally developed by Google.

“We have seen the importance that Istio has to complement the rest of the cloud-native ecosystem…[including] Kubernetes, Knative and Envoy,” Chen Goldberg, Google Cloud’s vice president of engineering and leader of the cloud provider’s open-source work, told Protocol in an exclusive interview. “We feel it will make it easier for those communities to work together under the umbrella of the CNCF, which is the home of much of the cloud-native technologies today.”

First released in 2017 by Google, IBM and Lyft, Istio extends Kubernetes to establish a programmable, application-aware network using the Envoy service proxy. It provides a uniform way to connect, secure, manage and monitor microservices, an architectural approach to software development that allows developers to separate applications into smaller independent parts, making them faster to develop and easier to scale.

The news about its transfer to the CNCF is being announced today at IstioCon, the community conference for Istio that runs through Friday.

The Istio project reached v1.0 in July 2018, when it was being used at scale by eBay and The Weather Company. More than 20 providers now offer Istio as a managed or hosted service. Those services include Google Cloud’s Anthos Service Mesh, a set of tools that helps customers monitor and manage a reliable service mesh on-premises or on Google Cloud.

“It's been a long journey with Istio, and we're really proud of what the community has achieved with this project,” Goldberg said. “We see an amazing adoption by our customers and users throughout the world with different industries, and we believe this is an important step forward. It will allow us to continue the journey of extending Istio's ecosystem and really amplify the affinity and the importance Istio has as a component in the cloud-native ecosystem, in particular Kubernetes.”

Google, which initially released Kubernetes in 2014, handed it over to the CNCF soon after the foundation was formed in 2015. Google had continued to manage and fund the Kubernetes CI/CD processes, container downloads and other services such as the Domain Name System until 2018, when it finally ceded day-to-day operational control of the project to the CNCF and Kubernetes community. In March, the CNCF also took official control of the Google-developed Knative open-source project.

“One of the things we've learned at Google over time is software and capabilities really thrive at scale when you wrap ecosystems and communities around them,” Will Grannis, Google Cloud’s chief technology officer, told Protocol.

This, however, was not always Google’s mindset toward Istio. Industry sources told Protocol in late 2019 and early 2020 that Google had blindsided its partners by alluding during the early days of development that it would eventually submit Istio to a neutral foundation, only to backtrack on that promise amid internal debate over whether the decision to cede control of Kubernetes was a mistake.

It would appear that open-source advocates within Google Cloud just won an argument.

“You can look at the legacy of Google and our most successful products – whether on the consumer side, or even now with our cloud platform – once you get a community active and once you get the ecosystem active around it, it really takes off,” Grannis said.

Google’s Istio announcement comes almost exactly two years after Google Cloud CEO Thomas Kurian told Protocol that the technology giant planned to donate Istio to a foundation following the confusion about its plans for the project.

Google was a founding member of the CNCF and continues to support the foundation as partners and contributors, “so that seemed like the perfect home,” Goldberg said.

“I acknowledge that there's been some confusion about the intent and the journey, but frankly, what we have shared also in the past is that we wanted to achieve a certain level of maturity in the technology and in the community,” she said.

The first step toward that was improving the quality and predictability of Istio’s technology with investments in automation and testing. In 2020, Google began to focus on three other areas of investment: the simplicity and ease of use of Istio, its extensibility and interoperability and evolving Istio’s governance model.

“Today, the Istio community has, of course, a working group, a steering committee and a clear path for maintainers with contributions to become leaders in the community,” Goldberg said. “We feel that we have achieved this important milestone, and we would like to further this engagement with the community.”

Are you keeping up with the latest cloud developments? Get the Enterprise team's newsletter every Monday and Thursday.

Your information will be used in accordance with our Privacy Policy

Thank you for signing up. Please check your inbox to verify your email.

Sorry, something went wrong. Please try again.

A login link has been emailed to you - please check your inbox.

Donna Goodison (@dgoodison) is Protocol's senior reporter focusing on enterprise infrastructure technology, from the 'Big 3' cloud computing providers to data centers. She previously covered the public cloud at CRN after 15 years as a business reporter for the Boston Herald. Based in Massachusetts, she also has worked as a Boston Globe freelancer, business reporter at the Boston Business Journal and real estate reporter at Banker & Tradesman after toiling at weekly newspapers.

The sector’s corporate venture funds won’t be sitting on the sidelines in a crypto winter.

The crypto industry has seen crypto winters before, but this one is different.

Biz Carson ( @bizcarson) is a San Francisco-based reporter at Protocol, covering Silicon Valley with a focus on startups and venture capital. Previously, she reported for Forbes and was co-editor of Forbes Next Billion-Dollar Startups list. Before that, she worked for Business Insider, Gigaom, and Wired and started her career as a newspaper designer for Gannett.

The crypto industry has seen crypto winters before. The price of bitcoin falls, investors pull back as they retreat from losses and startups go lean (or go under) as they try to survive. A major difference with this crypto winter emerging in 2022 is that the companies feeling the frost are also the ones doing the investing.

The last few years saw crypto companies play an active part in the crypto investing explosion. FTX Ventures put aside a $2 billion fund, and Binance raised $500 million to back the next wave of blockchain startups. The challenge now is that, as many companies cut their staff and trim costs, crypto corporate venture capital will have to balance taking advantage of opportunities in a crisis while navigating the pressure to conserve cash. Investors that touted the upside of access to a corporate balance sheet are now feeling the downside pinch.

“The budget going to the venture arm of these corporates is likely going down,” said Oppenheimer analyst Owen Lau, who covers companies like Coinbase. “It’s just hard to find an argument when they’re downsizing that you’re allocating more money into this venture, which typically invests in more longer-term-duration assets and does not generate immediate return for a company.”

It’s a tricky balance between short-term cash needs and the long-term strategic plays that can end up being the “hidden value” in companies like Coinbase, as Lau calls it.

The crypto exchange has one of the most active corporate VC arms across the tech industry, at times doing more deals than well-known Big Tech firms like GV (one of Alphabet’s funds) and Salesforce Ventures. But Coinbase has also been one of the hardest hit in the last few weeks. The company went from instituting a hiring freeze to rescinding offers to new hires. On Tuesday, it announced it was letting go of 18% of its staff.

Coinbase wasn’t spared in the layoffs with at least one team member posting publicly that they were no longer working with the company. (Coinbase declined to comment on the number of folks affected on the team.) But it doesn’t mean that Coinbase will no longer invest or that its budget has been majorly slashed.

“Coinbase Ventures does exist and is still very much alive. We just had our investment committee meeting today,” said Coinbase Head of Corporate Development and Ventures Shan Aggarwal when asked if it also fell victim to cost-cutting plans. “Despite the market conditions and market turmoil, we’re still very much continuing to invest.”

That doesn’t mean it's full steam ahead for Coinbase or anyone like in the market heyday of 2021. The deal pace has slowed down, both because founders don’t need a ton of capital, having raised a ton of money, and because some investors are tightening their belts and raising the bar of what they’re going to invest in, Aggarwal said. Crypto VCs are also in a stage of price discovery as they figure out what the new normal is going to be — which is harder to do when there aren’t many public market comps out there.

The upside of a pause for price discovery is the widespread belief that this is actually a great time to find deals. At the Consensus conference last week, the mood was still buoyant, with no one even saying “crypto winter” (out loud, at any rate).

Whether it’s a crisis or a bump in the road, the market meltdown and uncertainty has some companies taking advantage of the time to go deeper in investing.

Binance Labs just raised a $500 million fund in early June, including from outside investors like Breyer Capital and DST, to fund more blockchain and Web3 companies. The investment group is looking to take advantage of the price uncertainty to get better returns, Binance.US CEO Brian Shroder told Protocol’s Benjamin Pimentel.

“Frankly speaking, Binance Labs is in a really great spot right now because the valuations for a lot of these firms will be depressed given the crypto winter, and the ones that survive this winter will thrive in a bull market period,” Shroder said. “So to the extent that investors are actually investing now, that actually has kind of the greatest amount of return from my perspective.”

Coinbase actually got its start in the last crypto winter in early 2018 when it started investing in companies like OpenSea. Aggarwal isn’t fond of reminiscing on the price swings of that time — when bitcoin fluctuated from nearly $20,000 to $3,000 in 2017 to 2018 — but the turmoil helped investors like Aggarwal find the true believers.

“I'll sound like the ‘Oh, I've been in the space for a long time’ old guy at this point, but I look back on those days and I think I cherish them a lot, because from an investing perspective, the signal-to-noise ratio was a lot higher,” Aggarwal said.

History could repeat itself this crypto winter if other firms follow Coinbase’s path. In a pullback, less-competitive deals means there could be room in the cap table for other, smaller crypto companies and new investors, said PitchBook crypto analyst Robert Le.

What makes the crypto CVC space unique in the broader VC ecosystem is that there’s much more of an openness and a willingness to accept money even from a competitor. That makes the barrier to entry for a corporate VC even lower.

“The whole ethos is a rising tide floats all boats,” Le said. “Because we're at such an early stage of the development of the crypto space, there is acceptance of new projects accepting capital from Coinbase, even though they're developing a competitive product. I think once the market matures, you’ll see less of that, but right now you’re seeing a lot of investing in competitors.”

The question now is whether the tide can rise if the ocean is frozen over.

PitchBook’s Le expects crypto CVC to fall back roughly in line with broader VC investing. Unlike with the last crash, which a lot of CVCs sat out, Le doesn’t think any firm will choose to sit on the sidelines this time. The wild card here is that, unlike the 2018 crypto winter, this one coincides with markets tightening, rising inflation and widespread economic uncertainty.

“The broader implications of that on the funding environment are going to be very different. It could deter many founders from starting companies if they don't feel that there's sufficient investment capital for their business, and that's something that we're still trying to get our heads around,” Coinbase’s Aggarwal said. “It's very dynamic, and it's very fluid.”

Biz Carson ( @bizcarson) is a San Francisco-based reporter at Protocol, covering Silicon Valley with a focus on startups and venture capital. Previously, she reported for Forbes and was co-editor of Forbes Next Billion-Dollar Startups list. Before that, she worked for Business Insider, Gigaom, and Wired and started her career as a newspaper designer for Gannett.

Some of the most astounding tech-enabled advances of the next decade, from cutting-edge medical research to urban traffic control and factory floor optimization, will be enabled by a device often smaller than a thumbnail: the memory chip.

While vast amounts of data are created, stored and processed every moment — by some estimates, 2.5 quintillion bytes daily — the insights in that code are unlocked by the memory chips that hold it and transfer it. “Memory will propel the next 10 years into the most transformative years in human history,” said Sanjay Mehrotra, president and CEO of Micron Technology.

A kind of semiconductor, memory chips are integrated circuits made of millions of capacitors and transistors that can store data. They are the backbone of the broader digital economy — enabling everything from safer transportation to greater broadband access and a more efficient energy grid — and they play an active part in almost every aspect of our daily life. Consider automobiles, for example, “Not long ago, automobiles contained very little memory to process data,” Mehrotra said. “Today, automobiles process huge quantities of fast-moving data in support of advanced safety features.” Electric vehicles, in essence, can be likened to data centers on wheels.

The quantity of data has broader implications

The 21st century runs on chips, but right now they can't be supported. A global shortage of semiconductors, which crimped the production of everything from pickup trucks to PlayStations, has industry experts and policymakers looking for a new strategy to avoid debilitating bottlenecks in the future supply chain of microprocessors.

Semiconductors are in short supply because of several colliding factors, including spikes in demand for consumer electronic products and concurrent slowdowns in chip production caused by the pandemic. In 1990, the U.S. produced 37% of the world’s semiconductors. Today, it produces 12%. Meanwhile, 75% of the world’s chip manufacturing is concentrated in East Asia. Micron is the only memory manufacturer remaining in the Western Hemisphere, and it makes 2% of the global memory chips in the U.S.

The increasing performance demands caused by technological leaps in artificial intelligence, 5G, the industrial Internet of Things, edge computing, autonomous cars and high-performance data centers have increased demand for chips. Every major industrial company is trying to figure out how it can manage the amount of data it generates to make its business stronger. Chips are an essential part of that.

Meanwhile, foreign countries see the strategic importance of semiconductors and use government incentives to advance manufacturing while the United States government has watched it happen while sitting on the sidelines.

Take China, which is projected to have the world’s largest share of chip production by 2030 due to an estimated $100 billion in government subsidies. How much has the United States invested in essential strategic manufacturing operations like semiconductors? Zero. And, without government incentives, U.S. domestic chip production will continue to collapse.

After years of inaction, U.S. elected leaders have taken note, and are working to implement change. The House and Senate are working on legislation designed to increase global competitiveness, including the Facilitating American-Built Semiconductors Act, which aims to provide firms a 25% tax credit toward the construction, expansion or modernization of semiconductor fabrication plants and processing equipment in the U.S.

In January 2021, Congress passed the Chips for America Act. This bipartisan legislation authorized a series of programs to promote the research, development and fabrication of semiconductors within the United States. Together, the Fabs and Chip Acts are meant to renew American competitiveness in chip manufacturing, processing, packaging and design.

Large-scale investments like those the semiconductor industry are poised to make, coupled with targeted government incentives, can meaningfully benefit the U.S. economy in both the short and long term by boosting domestic manufacturing, enhancing supply chain resiliency and increasing national security. “Government support of these capital-intensive, long-term investments serve as a multiplier for advanced manufacturing across the industry, positioning the U.S. as a central figure in the fourth Industrial Revolution,” Mehrotra said. An investment in semiconductors is an investment in the foundational technology that will support innovation across sectors and industries in the United States.

Already there are hopeful signs. A billion-dollar chip factory just opened in upstate New York, and in October 2020, the Department of Defense awarded more than $197 million to help the American microelectronics companies create state-of-the-art design and manufacturing facilities. Advanced computer chips not only drive economic and scientific advancement but military capabilities. Through the Rapid Assured Microelectronics Prototypes plans, the DoD hopes to create a reliable and resilient domestic source of chips for its artificial intelligence, 5G communications, quantum computing and autonomous vehicle needs. “The microelectronics industry is at the root of our nation’s economic strength, national security and technological standing,” said Michael Kratsios, former acting undersecretary of Defense for Research and Engineering.

Ultimately, the result of the silicon resurgence — in processing power and speed as well as energy efficiency — will be to help build bold new business models, many of which we have yet to dream up. But, with the right investments in our chip-driven future, these dreams can turn into an exciting new reality and protect the future of America.

Important nuances were lost in translation when a rule commonly used to measure disparate impacts on protected groups in hiring was codified for easy-to-use tools promising AI fairness and bias removal.

Kate Kaye is an award-winning multimedia reporter digging deep and telling print, digital and audio stories. She covers AI and data for Protocol. Her reporting on AI and tech ethics issues has been published in OneZero, Fast Company, MIT Technology Review, CityLab, Ad Age and Digiday and heard on NPR. Kate is the creator of RedTailMedia.org and is the author of "Campaign '08: A Turning Point for Digital Media," a book about how the 2008 presidential campaigns used digital media and data.

Salesforce uses it. So do H20.ai and other AI tool makers. But instead of detecting the discriminatory impact of AI used for employment and recruitment, the “80% rule” — also known as the 4/5 rule — could be introducing new problems.

In fact, AI ethics researchers say harms that disparately affect some groups could be exacerbated as the rule is baked into tools used by machine-learning developers hoping to reduce discriminatory effects of the models they build.

“The field has amplified the potential for harm in codifying the 4/5 rule into popular AI fairness software toolkits,” wrote researchers Jiahao Chen, Michael McKenna and Elizabeth Anne Watkins in an academic paper published earlier this year. “The harmful erasure of legal nuances is a wake-up call for computer scientists to self-critically re-evaluate the abstractions they create and use, particularly in the interdisciplinary field of AI ethics.”

The rule has been used by federal agencies, including the Departments of Justice and Labor, the Equal Employment Opportunity Commission and others, as a way to compare the hiring rate of protected groups and white people and determine whether hiring practices have led to discriminatory impacts.

The goal of the rule is to encourage companies to hire protected groups at a rate that is at least 80% that of white men. For example, if the hired rate for white men is 60% but only 45% for Black people, the ratio of the two hiring rates would be 45:60 — or 75% — which does not meet the rule’s 80% threshold. Federal guidance on using the rule for employment purposes has been updated over the years to incorporate other factors.

The use of the rule in fairness tools emerged when computer engineers sought a way to abstract the technique used by social scientists as a foundational approach to measuring disparate impact into numbers and code, said Watkins, a social scientist and postdoctoral research associate at Princeton University’s Center for Information Technology Policy and the Human-Computer Interaction Group.

“In computer science, there’s a way to abstract everything. Everything can be boiled down to numbers,” Watkins told Protocol. But important nuances got lost in translation when the rule was digitized and codified for easy bias-removal tools.

When applied in real-life scenarios, the rule is typically applied as a first step in a longer process intended to understand why disparate impact has occurred and how to fix it. However, oftentimes engineers use fairness tools at the end of a development process, as a last box to check before a product or machine-learning model is shipped.

“It’s actually become the reverse, where it’s at the end of a process,” said Watkins, who studies how computer scientists and engineers do their AI work. “It’s being completely inverted from what it was actually supposed to do … The human element of the decision-making gets lost.”

The simplistic application of the rule also misses other important factors weighed in traditional assessments. For instance, researchers usually want to inspect which subsections of applicant groups should be measured using the rule.

To have 19% disparate impact and say that’s legally safe when you can confidently measure disparate impact at 1% or 2% is deeply unethical,

Other researchers also have inspected AI ethics toolkits to examine how they relate to actual ethics work.

The rule used on its own is a blunt instrument and not sophisticated enough to meet today’s standards, said Danny Shayman, AI and machine-learning product manager at InRule, a company that sells automated intelligence software to employment, insurance and financial services customers.

“To have 19% disparate impact and say that’s legally safe when you can confidently measure disparate impact at 1% or 2% is deeply unethical,” said Shayman, who added that AI-based systems can confidently measure impact in a far more nuanced way.

But the rule is making its way into tools AI developers use in the hopes of removing disparate impacts against vulnerable groups and detecting bias.

“The 80% threshold is the widely used standard for detecting disparate impact,” notes Salesforce in its description of its bias detection methodology, which incorporates the rule to flag data for possible bias problems. “Einstein Discovery raises this data alert when, for a sensitive variable, the selection data for one group is less than 80% of the group with the highest selection rate.”

H20.ai also refers to the rule in documentation about how disparate impact analysis and mitigation works in its software.

Neither Salesforce nor H20.ai responded to requests to comment for this story.

The researchers also argued that translating a rule used in federal employment law into AI fairness tools could divert it into terrain outside the normal context of hiring decisions, such as banking and housing. They said this amounts to epistemic trespassing, or the practice of making judgements in arenas outside an area of expertise.

“In reality, no evidence exists for its adoption into other domains,” they wrote regarding the rule. “In contrast, many toolkits [encourage] this epistemic trespassing, creating a self-fulfilling prophecy of relevance spillover, not just into other U.S. regulatory contexts, but even into non-U.S. jurisdictions!”

Watkins’ research collaborators work for Parity, an algorithmic audit company that may benefit from deterring use of off-the-shelf fairness tools. Chen, chief technology officer of Parity and McKenna, the company’s data science director, are currently involved in a legal dispute with Parity’s CEO.

Although application of the rule in AI fairness tools can create unintended problems, Watkins said she did not want to demonize computer engineers for using it.

“The reason this metric is being implemented is developers want to do better,” she said. “They are not incentivized in [software] development cycles to do that slow, deeper work. They need to collaborate with people trained to abstract and trained to understand those spaces that are being abstracted.”

Kate Kaye is an award-winning multimedia reporter digging deep and telling print, digital and audio stories. She covers AI and data for Protocol. Her reporting on AI and tech ethics issues has been published in OneZero, Fast Company, MIT Technology Review, CityLab, Ad Age and Digiday and heard on NPR. Kate is the creator of RedTailMedia.org and is the author of "Campaign '08: A Turning Point for Digital Media," a book about how the 2008 presidential campaigns used digital media and data.

The recourse offered to consumers when their crypto is held by entities that are in trouble is slim to none, but that doesn’t mean businesses are getting away with it.

There are few protections for crypto investors.

The chaos that has wiped out tens of billions of dollars in crypto wealth has brought the issue of investor protections to the forefront. Consumers promised high returns and safe investments lost big in the UST-luna collapse, which in turn has caused DeFi entities like the Celsius Network to freeze withdrawals. Alongside the general crypto market downturn, the damage is tangible, giving urgency to efforts to regulate crypto with basic guardrails.

Celsius attracted lots of consumers because of the yields it offered, which were higher than traditional bank accounts. After Celsius paused withdrawals, swaps, and transfers between accounts last Sunday due to “extreme market conditions,” it left investors anxious about how and when access to their assets stuck in Celsius might be restored.

Regulators have been thinking about these issues for a while, but for many of the new investors drawn to crypto recently, the questions of what happens to crypto held hostage by an exchange or a DeFi lender are freshly puzzling.

In the context of DeFi lenders like Celsius, it’s unlikely consumers will be able to access their crypto deposits unless they are unfrozen by the lenders themselves. How existing investor protection laws might apply to digital assets is largely untested, and the global and decentralized nature of many DeFi entities makes it hard to figure out where to start in pursuing a claim.

“A lot of people have losses who didn’t understand the risks they were taking,” former Commodity Futures Trading Commission chairman Timothy Massad told CoinDesk, stressing that there is a need for stronger regulation especially for lenders in the crypto industry.

SEC Chair Gary Gensler has echoed similar sentiments. He highlighted how crypto markets promise high returns and don’t “give a lot of disclosure” at an event Tuesday: "I caution the public. If it seems too good to be true, it just may well be too good to be true."

Even centralized exchanges can seem unsteady given the unsettled state of crypto regulation. After Coinbase added a new disclosure in its 10-Q SEC filing in May, it caused another wave of anxiety in the crypto community.

“In the event of a bankruptcy, the crypto assets we hold in custody on behalf of our customers could be subject to bankruptcy proceedings and such customers could be treated as our general unsecured creditors,” the filing said, which wasn’t a good look on Twitter.

Chief legal officer Paul Grewal quickly clarified that the new disclosure was only to comply with new guidelines issued by the SEC, and that there would never be a situation where “customer funds could be confused with corporate assets.” Coinbase updated a user agreement to reflect that customer assets would be protected under Uniform Commercial Code Article 8.

That UCC provision is one of the few rules lawyers might try to use against crypto entities, because the usual protections that cover deposit or brokerage accounts are tough to apply. Crypto doesn’t have protections like FDIC insurance has for traditional bank accounts or SIPC coverage for brokerage accounts.

For entities based overseas, investor protection laws vary from jurisdiction to jurisdiction, but because of how new the crypto industry is, many jurisdictions have yet to come up with laws to safeguard crypto assets for consumers.

But risky behavior by crypto companies isn’t flying under regulators’ radar. A look at the ongoing investigations Terraform Labs is under after the UST-luna stablecoin collapse gives an idea of the repercussions a DeFi company can face from causing losses for investors.

In one case that predates the current crypto crash, the SEC triumphed over Terraform in establishing its regulatory jurisdiction. A judge ruled that even if Terraform is based overseas, because it was in business with U.S.-based consumers, investors, employees and entities, it could be sued here. Investigations were also opened in Terraform founder Do Kwon’s home jurisdiction, South Korea, alleging losses for about 280,000 citizens.

U.S. state regulators have also stepped up in the Celsius case. Securities regulators in Alabama, Kentucky, New Jersey, Texas and Washington are reportedly probing Celsius about the withdrawal and transfer freeze, stating that the investigation was a “priority” because of concerns over consumer access to financial accounts.

With the introduction of the Responsible Financial Innovation Act in Congress and Biden’s executive order outlining how federal agencies should be working together for a regulatory framework for the crypto industry, federal legislation and regulations could be expected in the next few years.

In the meantime, SEC Commissioner Hester Peirce, a frequent critic of the agency’s scant rulemaking on crypto, gave one piece of advice to consumers back in May. “I think no matter what you're doing with your money as an individual, you need to be using your own brain, first and foremost. You need to be looking out for red flags,” she told Protocol.

Expanding from tech startups to mom-and-pop shops created a dilemma for the financial software company. The result: a painful decision to “offboard” many of the smaller customers it had recently courted.

Brex co-CEO Henrique Dubugras has had to "offboard" customers as the company focuses in on VC-funded companies.

Benjamin Pimentel ( @benpimentel) covers crypto and fintech from San Francisco. He has reported on many of the biggest tech stories over the past 20 years for the San Francisco Chronicle, Dow Jones MarketWatch and Business Insider, from the dot-com crash, the rise of cloud computing, social networking and AI to the impact of the Great Recession and the COVID crisis on Silicon Valley and beyond. He can be reached at bpimentel@protocol.com or via Google Voice at (925) 307-9342.

It’s been a rough week for Brex co-CEO Henrique Dubugras as he dealt with the fallout from a business fumble.

Brex had sent emails to tens of thousands of small businesses, telling them that the financial services company would no longer be able to serve their needs. After expanding its business from tech startups to traditional small businesses, including mom-and-pop shops, Brex had decided to pull back to its original core customers.

But the emails led to confusion, sparking harsh criticisms online. “This Brex account closure sucks,” one Twitter post read.

“It’s obviously a tough, painful day,” Dubugras said.

He explained what happened in an interview with Protocol, discussing why Brex had first moved to expand its reach to more traditional businesses and why it eventually decided that it had to withdraw from a “huge” market.

This interview has been edited for brevity and clarity.

Start with telling us what happened. Some people interpreted this as a move away from startups.

Let me share a little bit of historical context. We started a company in 2017 focused on serving startups. We could underwrite them based on cash balances. We gave them a credit card based on it. It did super well.

Then in late 2019, early 2020, we’re like, “OK, how do we expand from here? What's the next phase of products?” Brick-and-mortar small businesses seemed like a good way to go. So we built a lot of our systems to be able to onboard them.

I'd say we were pretty surprised by the sheer amount. There are tens of thousands of startups in the U.S. versus tens of millions of small businesses. The scale that that took was very, very big. We thought it was going to be fine; we'll just invest more to give them exceptional service.

At the same time, there was another effect that was happening. Our core customers, the startups, they were starting to grow. As they grew, they started having all these new needs. They’re like, “Look, we need you to solve these new needs that I have around spend management and global [expansion].”

What we realized was we couldn't do both at the same time. We couldn't serve millions of small businesses around the U.S. and create products for the needs of our best and growing companies.

We made the painful decision to exit that kind of traditional, small brick-and-mortar business in order to focus on startup businesses. Our startup customers require us to be able to grow with them for a longer period of time.

How do you define a startup and the businesses you’re planning to continue serving?

It's not a perfect definition. Our definition is anyone who received any kind of funding from either venture capital, angels, accelerators, any kind of professional funding. That’s the startup that we remain deeply focused and committed to.

These are mainly tech startups, right?

How big did the traditional SMB segment grow for your business?

I would say that the amount of companies that we onboarded every month multiplied by 25. So think of that and how that impacts a company.

What typically are these companies? Restaurants or retail shops?

Restaurants, retail shops, bakeries, florists, hairdressers, small design agencies. Small professional services, two-people design firms, things like that.

And if I am a business owner in those industries and was a customer, what do I have to do?

You need to move your bank account to a different provider.

You will no longer be serving my business needs.

Correct. Again, the reason we're doing this is so we can focus more on our core customer. We would love to be able to serve everyone and do a great job for everyone. But we made a tough choice of focusing on where we started.

What percentage of your total business will be affected?

I don't think we have any numbers to share there.

Are these hundreds of businesses, or thousands of businesses?

That we're offboarding? It's definitely in the tens of thousands.

Obviously, there’s been some confusion. Can you comment on how the plan was discussed and executed?

Yeah, absolutely. Look, it's something that honestly, for the longest time, we tried not to do. Our original plan was: We're going to do both. We as an organization are very capable. We have a lot of people. We have a lot of resources. We're just gonna pain it out and do both. Both are amazing markets. These are great business opportunities. We tried that for the majority of 2021.

Then by the end of 2021, it got to a point where we started questioning: What do we do from here? Do we sacrifice experience for our core customer? Do we allow our best customers to leave because we're not serving their needs? Do we build more products for everyone? Do we double the workforce? What do we do?

And this is the only solution that we could come up with. We weren't willing to sacrifice the quality of our service for our core customer. Especially in this macroeconomic environment, our core customer was pushing us to go even faster. They were saying, “Hey, I need to hire more people globally. Can you build more global stuff? I want to control more of my spend. Can you build more controls and more spend-management things?”

They were pushing us to go faster in a lot of things. It was just really hard to do both at the same time.

And we’re like, “We have to do this. We're going to do it once. So we're not going to start offboarding a little bit now and a little bit two months from now, a little bit three months from now. We're gonna do it all at once, one clean cut and make it very clear everyone knows where we're focused on.

On the execution, I would say that, probably if I were to go back, I would have been more clear about the distinction between startups and small businesses and what qualifies each one. Looking back, I still think it is the right decision for our core customer.

What did you mean, there should have been a clear definition?

Did we misclassify any company? Probably. It's a lot of customers. We're not perfect. If we figure out we made a mistake there and they do fit our definition, we will support that. So it is reversible. We will support them.

But that being said, I think that when we say small businesses, I think some people interpreted it as startups as well, which is very bad for us because we're doing this in order to support startups even better. That’s the complete opposite message that we were trying to send.

There are a lot of gray areas. You talked about design firms that could be serving tech startups.

That's why we use venture funding as the criteria. If any kind of professional investors invest in your company, that's our criteria.

It will be puzzling for some that you have all this demand, customers who want your service, and you're saying, “No, we can't serve you.”

The needs of these customers are actually quite different. It wasn't that they were asking us for the same thing, right? The startups were asking, “Hey, can you help us hire globally faster? Can you help me control my spend through software?” The smaller customers are asking, “Hey, can you give me a line of credit to weather the storm? Can you advance my receivables? Can you give me a lease financing?”

It was completely different needs.

But aren’t their needs, in a way, simpler? Why couldn't you sustain that segment of the business given the size of the SMB market?

It's huge, yeah. It's a great business. It's not simpler, actually. It's not more complicated, either. It's just different. When we're onboarding a startup, we can have white-glove service for them, talk to all of them on the phone, help them through everything. With a small business, it's not economical to do because there's so many of them. There's tens of thousands, even millions, so you need to have all your systems extremely automated, extremely perfect. You can't be hand-holding. Everything needs to be super, super scalable. We could get there eventually, but we have to invest a lot of resources in getting there right to be able to keep scaling.

Fintech lenders targeting traditional small businesses emerged because traditional banks were saying it’s too expensive to address their needs.

It's true. There are amazing companies that are focusing just on this. If you look at Square, their whole thing is doing this in a super scalable way that's cheaper. That's their business. Our business has a nuance. Our customers, they grow really quickly.

With Square, if their core customer is the restaurant or the coffee shop, they're not saying in two or three years, “OK, now we're Starbucks. I need all these new things.” Our customers in three years are like, “I need all these new things because I grew,” right? The fact that they grow makes us have to keep up with them.

The story now is you're abandoning the mom-and-pop shops, the restaurants, the retail stores and all those companies that make up a huge chunk of the SMB sector. How do you reflect on that?

You get this advice when you're a founder that focus is very important. When we started the company, we were 20 people, and we were like, “Hey, we built this product with 20 people. Why can't we just build all these other things with another 20?”

You think you can do all these things at the same time. I think that the reflection and the learning for me is you can do fewer things at the same time and you need to focus, otherwise you won't do either one or the other really well.

Again, it is really painful. Because we do understand the amount of stress that we're putting on a lot of small businesses, especially during this time. But we hope that you know people understand this is in order to serve our core customer.

And we wouldn't be able to serve these small businesses well because we're not building the new products that they need. And there's so many amazing companies out there and fintechs that their entire focus is serving them so they probably are better off betting on a partner that is focusing exclusively on that.

What are the next steps for you, given this change?

I think the most urgent thing is first reinforcing to our core customers that they are safe. We're not going to exit their market. And all of this was for them. That's probably No. 1. Second thing is being extremely supportive and using the majority of our resources over the next two months for the customers that do need to transition.

I wonder if you had any conversation with a restaurant owner or a retail shop owner or any small business owner who, during the pandemic, signed up with you and now you're saying, “We can't serve you anymore.”

I have, yes. And it's painful, because we did ask them to bet on us back then. And now we're offboarding them. So it's really painful for both us and for them. So we're super empathetic to it and we're going to do as much as we can to help them to transition.

Benjamin Pimentel ( @benpimentel) covers crypto and fintech from San Francisco. He has reported on many of the biggest tech stories over the past 20 years for the San Francisco Chronicle, Dow Jones MarketWatch and Business Insider, from the dot-com crash, the rise of cloud computing, social networking and AI to the impact of the Great Recession and the COVID crisis on Silicon Valley and beyond. He can be reached at bpimentel@protocol.com or via Google Voice at (925) 307-9342.

To give you the best possible experience, this site uses cookies. If you continue browsing. you accept our use of cookies. You can review our privacy policy to find out more about the cookies we use.