Business Models Startups Strategy

Robinhood Cash Management – daily tap, tap, go program

I don’t usually post about referral programs, but this one has to be the best I’ve seen in a while. Remember those early mobile games where to you had to frantically tap your screen to get more points or win the game? Guess what, product managers and UX designers at Robinhood did, too.

They launched the Cash management waitlist where you tap to go up in line. You can only tap for about 1,000 times per day, it seems. Pretty crazy, right?

Here’s what they did.

They launched Cash management, an added feature to Robinhood Financial LLC brokerage accounts. The Annual Percentage Yield (APY) paid by Sutton Bank is 2.05% as of October 8, 2019, and can be changed at any time. If you’re interested, use my signup link and we’ll both get rewarded.

Previously, they tried this thing before with savings accounts, but regulators shut them down, since they didn’t ask for approval beforehand. Hopefully this time it’s more legit!

Don’t forget to tap daily after you sign up!


The thing with Americans and their attitude

I met with a friend at the best coffee shop in town while I was in Baia Mare, visiting my family. She had just returned from Indonesia after years of traveling and living abroad. We both shared stories about how alien we felt in this town and how travelers always feel like this when surrounded by less traveled people. She told me about the people in Indonesia and the beauty of their culture, and then she asked me about the US, about life in California.

After delivering the classic picture of a tech manager in Silicon Valley, along with a few observations about the cultural differences between Eastern Europeans and Californians, I shared something that I found remarkable about people in San Francisco that I had encountered both professionally and in a friendly setting.

But first, a little context – I grew up in Northern Transylvania, attending public schools and being brought up in a very Romanian fashion. You were supposed to be great in school, get top grades, not get overtaken by your neighbor. Sports was secondary, but I still took up competitive basketball for awhile. The coach was tough, the team was demanding and there was little pleasure in playing, except for the winning moments and training with the girls’ high school team.

I dropped out of the competitive part of the sport mainly to pursue random teenage things like partying, smoking, drinking and being rebellious — for the duration of my high school years. It was only in London that I picked up basketball again, playing with startup co-workers and investment bankers at Canary Wharf’s Reebok gym every week. When I eventually left London for the US, I tried to find the same thing in San Francisco. A few months in, through friends, I found the basketball group that plays in the Presidio every weekend and I’ve been going there ever since.

You’re probably wondering why all this context, what’s the catch? It’s about basketball.

You see, I am not a professional player, nor am I a great amateur. I’m average, even considering my 6ft 6in advantage over others. When I was back in Romania, or even in the UK, playing and missing or failing to pass resulted in disappointment comments and bad looks from fellow players. You were expected to punish yourself with some down talk, too. That diminished the fun of the game, but supposedly kept your competitive spirit up and in the game. In San Francisco, I started playing, but something was missing. I would miss shots, fail to pass, but the only person talking down or frowning was me. Everyone else was trying to encourage me to go again, “good try”, “great shot”, “good look” – even though I knew and they knew I was way off. That made the game more enjoyable. Slowly, I started getting rid of the self talk that put me down on the basketball court.

But it’s not just in basketball. I’ve seen this type of behavior all over, where your peers, your bosses and your employees give you positive reinforcement and encouragement, even if things are going south. That, too, was a cultural shift from the mostly negative, sullen offices I had grew up in as a professional. This type of attitude not only makes things more fun and enjoyable all around, but push you to do more things, take more risks and ultimately grow — ironically the same goal of the other type of behavior I was used to, but significantly more effective.

This reflected in my writing too, because when I started doing it, it was more of a rebellious spout, a manifestation of my personality, in Romanian and very stream-of-consciousness, even with the more trade pieces.

I lost that when I moved to the UK, and then to the US, and it took me a long time to get it back, since I felt I had to “impress” my past readers in English and held back from publishing due to fear of looking like a fool. I forgot the fun in blogging. And it was this observation of how people react to mistakes in pickup weekend games that showed me I have a different public this time.

My friend asked me if I like America. I said yes, and used this basketball story. She understood why.

Photo by Braden Collum on Unsplash

Brain Interfaces

The road to a fully digital brain v1.0

Ever since I was a child, I had been thinking about death and ways to cheat it, to survive beyond the abyss. People do it with books, content, personas, legacies, children and other ways they want to be remembered. But that was never good enough for me. It’s too passive, you can’t reap the rewards of it, you can’t see how generations interact with your brand/legacy/offspring and you sure hell can’t contribute anymore.

Recently, I’ve been following the latest developments in brain-machine interface, including early research, invasive methods or visions of the future from companies like Neuralink. There are about 20 or so organizations that CB Insights has labeled as working on connecting us to machines.

See, the problem is our own interfaces are 50,000 years old, like our bodies. While they function well and work for the hunter – gatherer, they are becoming more and more stressed – both literally and figuratively – by what we are trying to achieve as a species. Imagine you didn’t have to speak, type or even see, hear or touch, but everything you experience happens directly in your brain.

That’s the future I would like to see unfold in my lifetime and it will lead us to merging with machines, using them as our vessels after our bodies no longer serve us, which is in about 70-80 years now. Sure, with modern medicine we might reach 120-130 years, but it’s hard to imagine more than that at this point.

So come with me on a little journey to find version 1.0 of the first fully digital brain.

Early brain computer interfaces

You can probably use Google to find the early days of brain-computer interfaces, so I won’t go to deep into what’s been done. These guys put together a good history of research done so far on connecting brains to machines. Spoiler, it often involved drills, cables and electrodes attached to the brain. Not fun, also not easy to survive long with that kind of invasion to your skull, due to bacteria and stuff.

I had a chance to talk about this with folks at a Brain Machine Interface event in San Francisco hosted by Berkeley researchers. Even posted on Instagram about it.

April 2018, somewhere in San Francisco

Everyone agrees we’re still early, but significant progress has been made. Like this team playing chess with their minds, that was recently documented on Arxiv.

Non invasive BMI like the company Facebook bought

It’s no secret Facebook wants to build a brain-machine interface. They have been working at it since the 2017 F8 announcement. It only seems natural, after years of collecting our most private thoughts and relationships, with our consent, that they would think of the best ways to get more share of our attention and brain. I know some people find this creepy already, but I am intrigued of the possibilities of actually finding likeminded people on Facebook with the help of these interfaces.

Ok, side note, I would be disappointed if you didn’t smile at my likeminded pun and that would make me question your understanding of this article.

But back to Facebook – they reportedly spent over $1B to acquire CTRL-Labs, a startup working on a such interface. Imagine combining the CTRL-kit (their device package) with Facebooks knowledge graph and with Oculus VR. Wouldn’t that be the perfect alternative world, where you think and you become part of it?

I would love to see some early prototypes and examples soon, so we can meet our parents from across the world in VR and not have to fly 7000+ miles every time we want to hang out. Time will tell.

This part of the journey only enhances our experience, and creates new ways to interact. It won’t actually mess with our cognitive abilities. Ok, maybe marginally, as we learn how to interact in the new world. The next part is where things get really exciting.

Brain functional enhancement with external processors/neural nets

This part is more future looking, so there will be less examples, more me trying to conceptualize what’s coming once our brains are truly connected to computers, not just through a wristband, but for real.

There’s already some headway done here, as well. Scientists at Graz University of Technology and UC Irvine have had breakthroughs in restoring motor functions of people with spinal injuries. It’s still early days, but it gives people hope.

Restoring functions after injuries is great and will help loads of people lead normal lives after otherwise life-altering accidents or illnesses. However, the next level is augmentation with other neural nets and external processors. The human brain performance varies significantly due to how you’re brought up, what you eat, how you learn, what you listen to, how chemicals flow inside it. What if you could leverage add-ons, like a new language skill neural net, or a neural net that does advanced algebra for you so you don’t have to go through college to learn it, or code. You would plug in the new net and transfer that exact pattern into your existing cortex or you could carry the add-on around with you and not have to rewrite your own brain. Feels a bit like Johnny Mnemonic, but I believe it’s possible in the next 20-30 years. After all, the brain and the way we think is very probabilistic, eerily similar to how neural nets understand and process information. Theoretically, by enriching your organic level neural nets with these engineered ones, you could jump through knowledge hoops faster.

I recently read a book called The Forever War, where the Taurans share a collective, yet centralized, knowledge server, where they upload and download their collective experiences as a sum of the parts – each tauran. That’s another way one could enhance one’s knowledge – know everything that everyone knows at a given time. It’s kind of having Google inside your brain and not having to use a computer. Imagine the possibilities.

Quantum neural net

While some people argue that we are all operating with quantum concepts in our brains – either consciously or unconsciously – by making scenario analysis and running risk-reward queries, it’s still quite discreet for the vast majority of people and rather slow even for those who can do it while being aware of it. For example, card counting, chess, these are applications of this type of scenario planning.

Imagine you could outsource this analysis to a quantum neural net, to a quantum computer that you can carry around in your pocket or even inside your skull. It could be powered by your bio-electricity and it could help you boost your scenario planning skills by running parallel computing at unthinkable speeds. This would accelerate decision-making, research, trials, investment, dating, virtually all life events.

The requirement for this is twofold – a great interface that’s non-invasive and a functioning quantum computer that adapts to whatever problem your brains throw at it, while still being portable.

Actually, I can also challenge the portability concept, because you can link up to it via LTE/5G/future mobile interfaces and not have to carry it around. It’s a single point of failure and centralizes the system a bit, but we’ll have to make do with that vulnerability for a while, if that’s what it takes.

Complementary brain/computer

I did spill over into this next step a little. This evolution of the quantum neural net can also store and reference existing information and experience it has collected throughout its interactions with humans. It’s basically a rudimentary version of a brain companion, which becomes complementary to the human brain.

It won’t be as vast and as versatile at the beginning, but it will be a step up – since it will be able to operate untethered, kind of like your Apple Watch without the phone nearby. This opens the door to creating mini-skill-bots, virtual brains that perform one function of your own brain really well repeatedly and consistently – like the UiPath RPA bots.

I can go a step further and say that a marketplace can be created for these bots and they can be paid in micro-transactions via the blockchain, as they complete tasks for you or for others that need that skill. They could even help you in your own job, create new kinds of jobs and replace meaningless work altogether.

Again, there’s the conversation here if we can host them on a personal device / implant or if they can be cloud-based or blockchain hosted and run decentralized, as a service. That would enable a whole new arbitrage market – where you buy computing and storage, train the mini-brain and sell its services to pay for the initial investment / running costs. Endless possibilities.

Fully digital brain on a server and/or distributed system

We’re now at a point where we’re not only able to communicate back and forth with machines, but we can also create mini-brains that run on prem or in the cloud, kind of like brain microservices, if you know what those are. No worries if you don’t, they have been explained on that page I linked above.

The challenge now is to connect those microservices and form a fully digital brain, that’s capable of operating on its own, without the human creator. It would have to be able to learn and decide on its own, contribute to society or get crushed by the debt incurred by its running costs. This brain, or should I say entity, is the singularity, the moment humanity creates new “life” inside a computer, life that can replicate, contribute, interact, learn, destroy, torture, you get my point.

This is now the point where we have a competitor on our hands – sapiens digitalus. I lost the human part of it for the sake of evolution, but who knows, it might be more human than many of us on this planet today.

Fully digital brain on blockchain systems for immutability

As we are building the world computer on multiple blockchains, it’s only natural to now think about hosting the new digital beings on blockchains, too, so they become immutable. If we host them on servers (prem or cloud), they are still bound by one single system, a single point of failure. While this vulnerability could keep them closer under human control, it could also limit their evolutionary capabilities.

What if we could release them in the blockchain and have them operate on the world super-computers? That way, they cannot be easily destroyed. Sure, you can already see Skynet rising and humans being driven to extinction, but this could also mean we can use them as vessels to preserve human knowledge, advance it and transcend with them.

A step up from here would be for the digital beings to run their own chains, build their own factories, robots and servers, produce their own energy and finally find ways to get out of the solar system to explore other worlds. For them, time does not matter anymore, since there is very little decay compared to the fragile human shell we currently call our body. They could travel for thousands of years at near light speed to jump from galaxy to galaxy and understand more about the universe that we would ever be capable of. In an ideal world, we’re friends with these creatures and they help us grow our collective knowledge. In a bad scenario, we’re extinct or kept in zoos/breeding planets, as a diversity and preservation effort. Wouldn’t that be ironic?

Conscience Immortality

I didn’t want to reveal my goal until the very end — but these digital beings that I mentioned above are likely to be us, the elders. My dream is for technology to advance to the level where we can upload our entire life experience as we approach an age where the body becomes inhabitable and we have to die. This is the equivalent to immortality. Sure, you won’t feel the same feelings, you won’t taste food, you won’t have the same type of experiences, but with tech advancements, you should be able to plug into a network of sensors far superior to what your body could offer and find new ways to feel the world.

Imagine having uploaded your brain before death and being onboard a starship that wanders through the galaxy exploring new life. Or being able to continue to live on, find new goals, connect with new people, see how your offspring and their offspring and their offspring do, talk to them and grow your circle for eons.

Hell, it could even become boring after a while and bring forth the digital turnoff movement, the digital beings/brains that have had enough and want to be shut down. But I doubt that such beings won’t find more purpose in a world that’s changing every day.

In summary, I’m excited about the world and what we can achieve a society, if only we would forget our little bickering and focus on technology advancements, AI, brain interfaces, body repair and restoration and improving the way we interact with our environment.

Photo by jesse orrico on Unsplash


I want to refocus this blog (again)

You stare at a blank page and you think — this is going to be my capstone, the best piece I have written so far. Then you start putting pressure on yourself to deliver and it becomes a chore, a burden. That dries up your inner writing flow and the blank page stays blank.

I have done that here. I have done that by boxing my writing into certain topics, trying to set fixed expectations when my mind is nothing but fixed. I want to change that now. 

10 years ago, when I first started writing, it was raw, unfiltered, unfettered, and it spoke to a few people. Then it spoke to more and more, soon becoming one of the best professional blogs in my country. 

But I moved to the UK in 2014 and part of me got left behind. I thought I could continue writing the same way, but I failed to recognize the fact that I couldn’t force a flow in a new language, a new industry and a world I had not experienced. 

After all, it took me 22 years to start writing in Romania, so why shouldn’t I expect at least some lead time? It only took me another 5 years to get to a point where I don’t need to box the writing in and let it flow. 

A few months ago, I wrote about how life is an improv game. Mine combines AI, blockchain, brain-computer interfaces, super-countries, social evolution and multiculturalism with a strong liberal flavor. 

If you are reading this and wondering what’s going on, look inside, let got and ride the flow. It will take you where you need to be.

For me, the show is just about to start. Act II. Yes, and…

Photo by Paul Skorupskas on Unsplash

Startups Strategy Venture Capital

Take the meeting after doing due diligence for VC expertise

I wrote an extensive article on how to choose the right VC as a founder after a few very direct and intense experiences and significant research. Recently, I have been advising a few founders and one of the most important topic they bring up is raising money. They have to go out and talk to VCs all the time, if they want their startups to grow. Sure, they could build and scale organically, and kudos to the ones that do, but for deep tech, more often than not, you need deep investments.

The first rule of VC meetings as a founder should be: don’t meet everyone who wants to meet with you. It’s the same as reaching out to VCs, not all of them will want to talk to you. And that’s ok.

Let’s say you are in the fortunate position where VCs reach out to you and ask for meetings. This is the way I would structure the due diligence process from the founder/CEO perspective:

  • Does the VC have experience investing in the market(s) I’m building for or with customers I want to have?
  • Does the VC have successful exists or meaningful M&A activity?
  • Does the VC have partners that have specific, hands-on industry knowledge that can benefit my business? (n.b. hands-on can mean operator, but it can mean a good track record as a VC in that space)
  • If you are deep tech, does the VC have deep technical expertise – like for NLP, blockchain, biotech, self-driving cars, manufacturing etc.
  • Is the person who reached out / who I’m supposed to be meeting the right person in the VC to meet?
  • Does that person have deep experience in my space – industry and technology?
  • Do they agree to take a 15-30 min phone call to discuss their questions?
  • Do they ask the right questions during that call that reflect industry and tech expertise?

If even one of these is a no, then you would be better off passing the call. They might be fishing in the wrong pond or just gathering market intelligence. It won’t likely lead to you getting investment from them.

If this is useful, please share with your founder peers.


Taxes around the world in 2019

My favorite topic: taxes

Having grown up in Eastern Europe and worked in the UK for 2 years, I know how painful it is to hand over around 50% of your gross monthly proceeds to the government. To add insult to injury, if you are a healthy, upstanding, self-sufficient, net-contributor citizen, you don’t even get to benefit from all those services you pay for every month.

In Romania, it’s easy to overlook all the social security and healthcare taxes, as they are drawn separately from income taxes and corporate taxes. They are, however, the biggest chunk of what the government takes from each salaried employee and every income-making resident. I won’t go into detail about how inefficient the system is with the ~50% they take.

The UK takes about 40% for mid-to-top earners and 45% for top earners. They offer much better services than in Eastern Europe, but the healthcare system has much to be desired, even with all the tax allocations.

The US has a more interesting system, with federal, state and local taxes. Even with all this stack, they don’t even come close to Romania or UK in terms of %. Sure, you pay for private healthcare and there are private schools and other non-state run services you’ll end up accessing. But you do that in Romania and the UK too, if you want top quality care, on top of taxes paid.

Check out this cool infographic and spot the facts that stand out for you. Maybe you can choose a country to live in based on this, or where to incorporate your next startup:

Source: Fortunly


Life as an improv game

Last year, I accidentally got myself and my wife into a walk-in improv class in San Francisco. It was one of those things that I had never done before due to cultural reasons and wanted to put myself out of my comfort zone by doing. We got there early, so we chatted with a few of the people there – some had done this before, others were new, just like us. The instructor then led us up there, in some old conference room belonging to the SF Chronicle. The instructor was a guy who worked as an Engineering Manager at Google and lots of the participants were techies, so we felt right at home.

The setup was not amazing, but we quickly forgot about it with a slew of games that made us more aware of each other, more in the moment.

More importantly, it showed me that I’m potentially good at this acting thing. I had thought the opposite my whole life. So I decided to double-down on it, by booking a full 6 week class of Improv I. If you’re in San Francisco, you can do it, too. It’s really easy, 3 hours every week, and if you miss a class or two, you can easily reschedule. To graduate, you need to attend at least 5 of the 6 classes.

I went there with an open mind and I got a lot more than I was expecting, hence this article. Improv, unlike standup comedy, is more about creativity and playing than it is about being funny or making jokes. Being able to play a scene and be consistent, authentic is what matters more than to be witty or crack the best punchline. Often, the best punchlines were the ones unplanned, the ones that came naturally as part of the relationship being played there.

I recently finished a book called “The Triangle of The Scene” – spoiler, it’s about improv. The author, Paul Vaillancourt, who’s an improv veteran, gives a little bit more structure around these plays and what makes them work – the relationships being built and played in the scenes

What I’m doing // What you’re doing // What we’re doing about it

Triangle of the scene

Reading this book and doubling down on my experience with the first improv class got me thinking about life and how we approach our relationships with others and with ourselves. There’s much to be learnt from the power of “yes, and” and the focus on relationships rather than the narrative. Try following your vocabulary at home and at work, see how many “yes, and”s you say and how many but’s, however’s, no’s and other negations we use on a regular basis.

I bet you’ll find the negative most often outweighs the positive. Try to flip that ratio for a week and see what happens. You may be very pleasantly surprised. Many leadership books and talks focus on getting people to prioritize positive relationships. Improv takes a different look at things.

Wish I had done this earlier in my life. Worth it and scheduling the next modules as soon as possible.


Tweaking the (Un)Fair Isaac and Company score aka FICO

When we first got to the US, we knew that the first thing we had to do is get a credit card and build credit history so we can actually have access to goods and services like real Americans.

This is thanks to the FICO score, a thing developed in 1989, after the FCRA legislation was enacted in the 70s. That’s when Equifax was also born, which later gave way to the biggest identity heist in history, with over half of US residents affected.

What’s this FICO score? It’s what banks, real estate companies, credit card companies and other service providers use to understand if you’re going to pay your bills so they can offer credit of some sort. In principle it’s a good thing, giving access to credit to a broad part of society and allowing everyone to play the game and work towards that dreamy perfect 850 score (which is not worth the effort, by the way, as you’ll do fine with a 780+ score).

Image Source

But coming back to my issue with it, it’s very rewarding to people who were born in the US and have had credit histories their whole lives, granted they have always been on time with payments and disciplined with their finances. Sure, the ones who overextend, miss payments and default will be hit. But the ones who play the game can win at it.

Except for immigrants. 15% of the score is length of credit, so an immigrant who just arrived to the US is already way behind in the credit game. It will take them 5-10 years to get to a point where they can hope at a perfect score. This is because FICO scores +10 years of history as the best category and removes points for anyone who has less than that. This is the unfairness that I mentioned.

It could be fixed easily. By adding the word “proportional” to the way credit history linked to the moment you become a lawful permanent resident or a lawful non-migrant alien is calculated in the FICO score, they could level the playing field and give access to better credit to more people, not rob them by asking for double digit car loan terms or extremely expensive mortgage rates just because they weren’t born here. 

What’s the reason for this absolute anchor? Why having 10+ years of credit is better than having credit 100% of the time you’ve been here? Why 2 years of history is worse than 10 years? Will FICO ever change? 

I have started making the argument that credit bureaus need to rethink their business models. In an era of identity theft, hoarding customer information and selling it is no longer a sustainable business model. I should be able to own my own score, based on my own accounts. It should be in my best interest to add as much information to a secure, local datastore and then selectively share proof of these transactions whenever I need to prove my creditworthiness.

Not the other way around, to have 3rd parties store my data and share my information freely with lenders so they can advertise all those loans in my mail.

It’s time for credit scores to change. 


The conservative and progressive irony of liberals and republicans

Thinking about AI, progress, innovation and pro-business can only lead me to think about politics. Since I’ve been in the US for a while now, I’ve started to spot some interesting anomalies, contradictions in the way the two dominating ideologies tend to behave conservatively or progressively.

Why does this matter? It’s all about double standards and the idiosyncrasies that people are so comfortable living with. Harari makes it his mission in Sapiens and Homo Deus to reflect on and uncover the vast array of contradictions our society deals with every day. You might think this is a new problem, but It’s not. It’s as old as religion, the main source of contradictions – like thou shall not kill, unless they are an infidel. You get the point.

Let’s start with Phoebe, who’s a dedicated, lifelong liberal. She lives in California, works in tech, and spends her free time hiking, doing yoga and going out to brunch. Pretty familiar, no? While she values progress, at least at declarative level, she opposes change in her neighborhood, wants to regulate everything that may impact any people negatively (which, by the way, is almost everything people do out there), but claims that protects freedoms and is progressive. Progress in her mind is to protect and conserve the status quo of any kind, while slowly innovating within a government set box. She doesn’t know it, but she’s conservative.

What about Lindsey? She’s a dedicated, lifelong republican. She lives in Louisiana, works in the insurance business, spends her free time at the local social club, watches TV, likes to drive to malls and organizes regular barbecues. Also familiar, no? While she’s very conservative, at least at declarative level, she’s pro business, wants less regulations and more individual freedom for people to do whatever they want within reason and without harming others. Conservatism in her mind goes back to the constitutional rights that were put in place by the founding fathers, while everyone today can do pretty much what they like and try to find new ways to create value, without the government dictating what people can or can’t do in their communities. She doesn’t know it, but she’s liberal.

This is the main source of contradiction in the American culture today, as both Phoebe and Lindsey want the same thing, to live the best life they can with their families, their friends and within their community. They can’t seem to see eye to eye on most things because they don’t know they share a lot of things, much more than what sets the apart. So if both of them are both liberal and conservative at the same time, it is actually easy for them to connect, if they set aside the outrage TV and politicians instill in both, setting them up to hate each other. 

This time I have no solution for this, other than making people more aware that they are not so different from their political counterparts. Food for thought for 2020.

Image Credits


Surviving as an AI in the US – the most litigious nation in the world

In the past few years, AI advancements have made self driving cars mare of a reality than the remote possibility it once was. Thanks to GPU advances, algorithm improvements and lots of venture capital thrown at the problem, Waymo was the first company to release its fleet (albeit in a very limited fashion) in Phoenix, AZ.


I was talking to my real estate agent there about this amazing feat, when he raised a very good point – how can companies that develop self-driving cars protect themselves from class-action lawsuits brought about by ethical dilemmas like the one where AIs have to choose to kill the driver over an oncoming car or a grandma over a baby. This is a delicate problem even for people, as we’ve seen in recent studies, where each culture determining the statistic outcome. 

We can’t decide for ourselves, so it’s unlikely we can ask the AI to do that too and make it acceptable for all cultures, with a blanket procedure. 

Here’s an example:

A self-driving car encounters a speeding human driver on the other side of the road swerving out of control. It evaluates the problem and sees that it has two options: it hits the other vehicle and throws it off a cliff, killing the other driver and its potential passengers OR it swerves out of the road and into the canyon, killing its own passengers. Regardless of what it does, the AI will kill someone. This, in the US and other jurisdictions, will likely cause the company operating the AI fleet to be sued and held responsible globally for an isolated incident that they couldn’t control anyway. Meanwhile hundreds of thousands of other self-driving cars operate normally.

This kind of blanket approach will likely cripple adoption and innovation at scale, due to the highly risky nature of driving today. Yes, you are far more likely to die in a car crash than with any other means of transportation known to man.

So could be a solution here? Treat AIs like franchises, once trained. Companies like Google, Uber, Lyft, Tesla and others could treat each AI, once trained, as independent entity, under a micro LLC, where the liability of that AI is limited to a number and to certain recourse. Whomever owns/operates it carries part of the liability, as to maintenance, configuration, potential fallback mechanisms, so that if an accident were to happen in Astoria, NY, it wouldn’t affect cars in Sacramento, CA. 

Right now, car companies don’t carry all the liability for their products, as long as they operate within expected parameters, as the decisions are made always by humans at the wheel. What if you could configure your AI when you buy the car to kill the grandma over the baby or protect the passengers over the other car? Then you would effectively release the AI and its parent of that liability and enable more innovation to be adopted faster.