Settings

Theme

Element AI sold for $230M, as founders saw value mostly wiped out: document

theglobeandmail.com

54 points by yazr 5 years ago · 58 comments

Reader

nrmitchi 5 years ago

The actual title is "Element AI sold for $230-million as founders saw value mostly wiped out". This is an important difference, since the submitted title here implies they were entirely wiped out.

It looks like this company raised ~$250M, and sold for $230M. This isn't any sort of nefarious "founders got wiped out"; they sold for less money than they raised and this is the typical outcome in that situation.

Also, from the article:

> As part of the deal – which will see ServiceNow keep Element AI’s research scientists and patents and effectively abandon its business – the buyer has agreed to pay US$10-million to key employees and consultants including Mr. Gagne and Dr. Bengio as part of a retention plan.

This is the kind of "everyone got wiped out but the founders got an out-of-band parachute" that highlights an often-understated difference between "founders" and "just an employee". Even with a terrible outcome, founders and top executives can still manage to come out on top, and only employees actually get wiped out.

  • dang 5 years ago

    Thanks—we've reverted the title to what the article says (shortened slightly to fit the 80 char limit).

    Submitters: "Please use the original title, unless it is misleading or linkbait; don't editorialize." https://news.ycombinator.com/newsguidelines.html

  • rossjudson 5 years ago

    If preferred shareholders gain control, common shareholders are screwed (as in zero value). I speak from personal experience. In our case, the founders and top executives absolutely went down with the ship and lost everything.

    Briefly, cash was needed and investors were found at an relatively early point. The agreement set a value (V) under which the proceeds would benefit the preferred shareholders (the new investors). After the agreement was signed, the new investors joined forces with another investor achieving control. A new entity was set up, controlled by the investors, and they used their control to force the sale of the company at V, instantly destroying all common shareholder value.

    All of this was announced (in a rather gloating manner) by the new head of the company. "You report to me now", basically.

    This did not sit well with the staff, who had been working hard for a while. I walked that day, and within a few days the entire engineering team did as well.

    For their $5 million, they got office desks, a bunch of PCs, and some half-built software they had no idea what to do with.

    I suppose that did not work out particularly well for anybody ;)

    • musicale 5 years ago

      > For their $5 million, they got office desks, a bunch of PCs, and some half-built software they had no idea what to do with.

      I suppose when they sold it off it was a good deal for some other startup that needed office equipment.

      > I suppose that did not work out particularly well for anybody ;)

      Sounds like the staff who quit probably enjoyed their revenge, and also got a great story out of it.

  • dmix 5 years ago

    Yeah this is basically part of the deal when you take VC and raise hundreds of millions. Don't expect much in return if you fail to deliver, or really unless you skyrocket at least 10x as expected in such arrangements.

    This is the sort of middle ground where it's not a complete disaster but some value is still generated out of the entity.

  • JoeAltmaier 5 years ago

    This always baffles me. You buy a startup that has made some progress. You want to keep going on the current product? Then keep the employees, let the 'founders' go. The founders have done their thing. But to keep the lights on will need everybody who knows how this unique thing works.

    Yet they always do the opposite. As if the founders have a clue how the software/deployment/support system works at all.

  • arolihas 5 years ago

    It also says ServiceNow agreed to hiring many of the employees in this deal so it’s not that catastrophic. Although yes their options aren’t even worth the paper they were printed on.

    • 908B64B197 5 years ago

      It's a real acqui-hire.

      They basically acquired a world class AI team. Had they let the company die I'm certain most of the AI talent there would have been poached rather quickly.

    • nrmitchi 5 years ago

      Ya, but let's not pretend that most of these employees couldn't haved walked outside and ended up with a full line up of interviews scheduled for the first week after the holidays.

  • Aeolun 5 years ago

    > only employees actually get wiped out

    Presumably they were paid a salary while they were working there.

    Having to find a new job now sucks, of course, but I find it hard to say they got wiped out.

agentofoblivion 5 years ago

I don't know anything about this company. But I work at FAANG as an ML scientist and spent about 3 months this year working on a research project that demonstrated big performance improvements of a new tech. It got enough interest that it led to a new project that brings this to production.

This has led to countless headaches. The skills needed to solve these challenges are very different. My more academically minded colleagues don't seem to have the intuition for seeing around this corner and understanding what it takes to build something that actually works.

This is a long way of saying: if a company is founded by and ran by such academics, I can see how this would be a recipe for disaster when it came to actually building and shipping products.

  • qeternity 5 years ago

    Really good insight. When I put engineering job posts up, we get loads of PhD applications...a staggering amount. And all of these people look great on paper. It’s only when you start talking to them off script that you understand a career in academia has radically different incentive pressures than the private sector. We end up passing on most, not because they aren’t brilliant, but because they’re very one dimensional, which is what suits academia.

    • musicale 5 years ago

      It must be surprising to find "brilliant" people who are apparently incapable of learning how to do new things.

      • jldugger 5 years ago

        I don't find it surprising at all, sadly. After a reorg, one of my colleagues on a different team (largely Master's degrees in ML) told our new team member: 'You shouldn't be worried, being assigned to their team is an honor. They're hardest to get into -- it's full of people who didn't just skip all the hard courses.'

        In many ways, hiring advanced degree holders is a crapshoot. They have skills you probably can't train, but often times come in with fewer software development skills than your undergrad intern, despite theoretically having more years of experience. You don't need to know git to publish in IEEE, or write unit tests or readable code, or debug an edge case, and your only code reviewer is a professor who doesn't care about this either. 'Good enough to publish' is a far cry from 'customers will pay for it.'

      • concreteblock 5 years ago

        I don't think it's fair to characterize such people as "incapable of learning how to do new things" if they're not even being given the opportunity to learn.

        I'm not saying that the OP should hire these people, or that they would even be good hires. Maybe they don't want to take the risk or time of training up a new employee.

        It just seems like a bit wierd non-sequitur to draw conclusions on someone's learning capabilities based on their performance in an interview outside their usual domain.

      • qeternity 5 years ago

        I never said they were incapable of learning new things. But they're usually not applying for junior level roles where we expect to make these investments. They are typically applying for senior roles where the applicant is expected to possess these soft skills.

        • concreteblock 5 years ago

          In your opinion, what are the important 'soft skills' that

          1) Academics lack

          2) Are hard to pick up

          E.g. another commenter mentioned Git. Many academics don't use version control, but I think learning how to do this to an acceptable level is not very difficult, so it doesn't really count.

          I'm asking because I might be trying to get a job in industry soon.

  • imafish 5 years ago

    I once worked at a consultancy that did government contracting. They had a hard time finding new software engineers to work as consultants.

    So they got the idea, that they could hire other STEM Master's and PhD's, offer them an average software engineer salary and train them up to become good software engineers. Many software engineers scoffed at the idea - so you're hiring people from Physics to do web apps - pfff, what do they know about creating software?

    Well, I'd say a third of my coworkers at the time were from "academia" - I felt absolutely no difference. The company is thriving and they have more than quadrupled in size since I worked there five years ago.

  • contingencies 5 years ago

    The skills needed to solve these challenges are very different

    Indeed. I founded a contactless personalized food production and automation robotics venture in 2016. We tore through specialist and generalist employees alike trying to make progress toward a final prototype. We're now essentially there, but lost the team over COVID, while the investment environment for our technology is currently extremely hot. In taking stock of the situation to move forward, I've realised it's actually a blessing: the skills needed to reach mass-production on a very complex assembly are very different to the skills needed to iterate early and mid-stage prototypes rapidly and effectively. It would have literally been harder - and more expensive - to drag people 'over the line' in to production than to just hire a new team.

  • JoeAltmaier 5 years ago

    TransMeta.

fierarul 5 years ago

Not sure what the angle should be here. The founders didn't manage to raise the company; it was soon to be bankrupt. Of course the founders would be wiped out.

Still, they have $300,000 worth of shares and "the buyer has agreed to pay US$10-million to key employees and consultants including Mr. Gagne and Dr. Bengio as part of a retention plan".

Business doesn't always work out. Getting a cool $1M after trying and failing is not the worst thing that could happen.

  • 908B64B197 5 years ago

    The company's value is in it's employees.

    They have a world class expertise.

    • fierarul 5 years ago

      That's not the market way of looking at it. How will a bankrupt company retain their world class employees? They are one missing pay-check and counteroffer away from moving elsewhere.

      • 908B64B197 5 years ago

        That's what I was saying.

        The value of Element AI was that by buying it you get all the employees. Had they waited for the company to go bankrupt they would be competing with everyone else trying to buy the IP and the key employees.

        • fierarul 5 years ago

          This I agree with. The only value remaining in the company are its employees, not revenue streams, not products, just employees and potential.

Barrin92 5 years ago

>Element AI invested heavily in hype and and earned international renown, largely due to its association with Dr. Bengio. It raised US$102-million in venture capital in 2017 just nine months after its founding, an unheard of amount for a new Canadian company, from international backers including Microsoft Corp., Intel Corp., Nvidia Corp., Tencent Holdings Ltd., Fidelity Investments, a Singaporean sovereign wealth fund and venture capital firms.

This has been bothering me for a while. Ridiculous sums of money invested purely on name recognition, in particular in 'AI', rather than allocating capital to people who actually at least have a product.

whoisjuan 5 years ago

Well, nothing surprising here. Element AI raised about 260MM, failed to create a revenue trajectory so it sold itself on the merits of its IP to ServiceNow for 230MM. Of course the founders and employees will get wiped out. Investors are first in line to get paid.

If you’re a founder and you ever are doing a big round try to seek for secondary liquidity. It’s fair for founders to have some way to build financial stability when they are sacrificing so much. Secondary liquidity terms are way common nowadays than they used to be.

Element AI founders should have done that when they did their Series B and raised 150MM.

I know there are some purists that think that giving founders secondary liquidity can remove certain accountability, but honestly if you’re investing in a group of people and you believe giving them some reasonable path to partial liquidity pre-exit, you probably have some fundamental doubts on that investment and that group of people. I feel that nowadays that’s basically a litmus test for investment strength.

ReDeiPirati 5 years ago

Unlike Landing.ai and C3.ai, Element.ai was lacking of a solid go-to-market strategy from day 1.

  • throwawayeai 5 years ago

    I worked at Element AI, and that was true from the inside to an extent that was borderline terrifying. On orientation we were told that EAI's strategy was to hire as many smart individuals as possible. There was no focus on delivering an actual product, it was demo after demo of semi-impressive DL models.

    It makes me sad thinking of all the brilliant colleagues I had that were simply wasting their talent.

    • amelius 5 years ago

      > It makes me sad thinking of all the brilliant colleagues I had that were simply wasting their talent.

      At least they were not making people click ads ...

    • gumby 5 years ago

      This was why jg bought DeepMind. Presumably they hoped lightning would strike twice.

    • rfrey 5 years ago

      That strategy worked out ok for DeepMind.

  • yourapostasy 5 years ago

    > Unlike Landing.ai and C3.ai...

    What do you believe the go-to-market strategy is of those two?

    For C3.ai, my interpretation was they're trying to become the Microsoft of ML, re-packaging the entire ML pipeline into a more consumable developer experience. They seem currently focused more upon the model analytics part of the pipeline than say, training data selection or ETL to ingest raw data (whether to train upon or run in production), or any number of other pieces of the pipeline.

    Landing.ai appears to be more tightly focusing their messaging on the operational aspects of the ML pipeline, though not so much the modeling part that C3 appears to emphasize. They also seem to very tightly narrow their focus on machine vision ML.

    I'm probably wildly off though, having had no access to the actual platforms and ever used them in anger.

    • shuckles 5 years ago

      I attended a presentation by Tom Siebel, founder of C3, around 2011 who said the "inspiration" from C3 came from hiring ~10 of the best management consultants from McKinsey who told him that Enterprise IoT was going to be a big deal. My guess is the AI part came later as a way of productizing a data processing pipeline.

      The event was strange and really disabused me of many assumptions I had about startup origin stories.

      • sjg007 5 years ago

        Enterprise IoT is going to be a big thing... You know because as Deloitte says to paraphrase, "your stuck on a plane that's going nowhere because of a technical.. turns out its a 10cent part.. how are you going to get there? You need a smart factory ecosytem, ... it's not a supply chain but a supply network" or however, the podcast ad goes.

        C3, Palantir, Salesforce and probably a few other companies / consults are investing in this space. It's the whole integrate all the data silos into some cloud, public or private, with +snowflake style data storage and build your business around data in order to drive real time feedback to solve customer problems.

        But you are not wrong, "productizing a data processing pipeline" is big business.

    • sjg007 5 years ago

      You should read the history of C3.ai to understand where it fits in the market.

      https://www.forbes.com/sites/alexkonrad/2020/12/09/billionai...

    • muzaffarpur 5 years ago

      My understanding is C3.AI primarily develop notebook extensions to work with data.

Apocryphon 5 years ago

So much for Montréal becoming the Silicon Valley for AI. Another Canadian/non-Californian startup bites the dust.

  • 908B64B197 5 years ago

    The problem is AI is a tool to build a product. It's not a product in itself.

    Element AI's product was it's talent (mostly all grads from MILA just next door) and the fact it's one of the few places where you could get as many great devs and deep learning experts in the same room.

    But as long as folks focus on AI and not building an ecosystem or a product we'll see these acqui-hires.

  • person_of_color 5 years ago

    Never 4get North

latenightcoding 5 years ago

Anybody familiar with the matter knows if Bengio scaled down his academic duties while raising money for Element AI?

  • mlthoughts2018 5 years ago

    He did scale them down, but he forgot to subtract the mean first which is why his duties were biased and the outcome was not efficient.

dvnguyen 5 years ago

Wow $200M funding for a startup with no products. Not even sure if they were serious in building a product or just cashing out the AI hype and academic credentials.

yazrOP 5 years ago

Paywall copy from reddit

https://old.reddit.com/r/MachineLearning/comments/khin4c/n_m...

legerdemain 5 years ago

In Firefox, turning on the reader view and reloading the page gets through the paywall, as it does with many paywalled publications.

mlthoughts2018 5 years ago

It’s funny to see this around the same time as this other thread where people are going through so much mental gymnastics to avoid obvious conclusions of how horrible it is to work for a startup,

https://news.ycombinator.com/item?id=25493646

  • brutus1213 5 years ago

    If the start up offers market-rate wages, I am not seeing the downside. Job insecurity vs. chance to get more experience/lottery ticket? Seems reasonable .. but then again, I don't work at a startup.

    • mlthoughts2018 5 years ago

      Start ups don’t offer market rate compensation. Typically salaries will lag by 30% or more compared with even mid-level tech companies, not even considering FAANG at all. That’s just base salary. Most startups don’t pay a bonus, and the equity is usually very poor on an annualized basis, and it’s mostly tied up in options where you have to deal with the strike price and tax issues.

      A wildly successful compensation outcome for an employee at a startup would be ~75% of a comparable market salary plus a few hundred thousand dollars of post-tax income from equity after 8-10 years, amounting to say $25,000 per year of equity/bonus.

      At a totally mid range tech company or large company with some technology, you would make 100% of that market rate and probably make $25,000 just in a bonus per year, plus an additional $30k to $50k of RSUs per year, which you don’t have to wait 8-10 years to sell.

      In an extremely conservative estimate, you’re probably losing $200,000 of post tax equity / bonus income over 8 years by choosing a startup over a mature but run-of-the-mill tech company or other large company that has tech teams.

      If your market salary is ~ $150k, the startup is probably paying $115k and will give worse raises and promotions. Conservatively that’s another $180k of lost post-tax income over 8 years (35k delta times 8 minus taxes).

      In total that’s about $400k of lost post-tax income by working at the start up for 8 years, and the alternative I’m comparing to is a very conservative estimate $150k base, $50k equity $25k bonus assuming no stock growth or raises, which is a totally run of the mill offer in large cities even 5 years ago for roles with 3-4 years of experience.

      And that’s under a huge IF the startup is wildly successful and all those options are actually worth a few hundred thousand after taxes. That’s a rare outcome. They could be worth nearly nothing, and then you’re talking about foregoing around $550k of post tax income over 8 years (225k all in per year minus 115k at a failing startup, per year, x8 then less taxes).

      Is the high, high risk of leaving $550k on the table worth it for vague promises of “interesting work” that isn’t guaranteed?

Keyboard Shortcuts

j
Next item
k
Previous item
o / Enter
Open selected item
?
Show this help
Esc
Close modal / clear selection