Kate Crawford: “I’m going to quote Paul Thomas Anderson here. I think, you know, we might be done with the past, but the past is not done with us. And as much as we might like to think that this is a new future technology, it is very much built on a range of both political and computational approaches, which will continue to be part of the fabric of quantum mechanics.”

David Mohring: “Feynman in the 80’s is often credited with bringing quantum computing to the more popular — if you call physics department popular — community. But it was really Shor, when he discovered this algorithm, that captured the imagination. It wasn’t the algorithm in and of itself so much that got people interested. It was that there was an algorithm. It wasn’t Shor alone. There were some other ones that are about that same time a little before by Deutsch that generated some interest. But what made Shor’s algorithm so interesting is it captured the imagination of the government. And the government saw this, obviously, with decryption. They started funding immediately. And Ion-Q co-founder Chris Monroe was at NIST and the group of David Wineland back in the mid-90’s, and they were one of the first ones to receive funding from the government. And one of my old colleagues was actually allowed to go there during this first kick-off program review, like analysis of how are you doing with the money that we’ve given you. He was allowed to stand up and say, I am so-and-so with the NSA and I’m allowed to say we’re interested in quantum computing. He sat back down. And of course, they’re interested in quantum computing. It has to do with cryptography. There’s no secret there. It’s in Wikipedia. It’s everywhere. And really, the intelligence community has funded quantum computing development for really that first 20 years. And it is slowly going up. If you look at the grand scheme of things, how much funding is going into quantum computing, obviously as a physicist, we’re very excited about the growth of the community. But if you compare it to many of the other activities that are funded by the government, it’s kind of dwarfed. I was part of a group that wrote a framework of the US government… Not just quantum computing but quantum technologies in general. It was published, I believe in 2016, it was about $200 million a year, which sounds like a lot. But if you consider like the overall budget of the of the US government is not relatively so large.”

Fleur Johns: “… I wanted to focus on the ubiquity of inequality in the jurisdictions and settings in which quantum technologies are largely being developed. Recent advances in technology have, I think, been built in vast inequality machines. That’s the case. Obviously. I’m talking metaphorically about a vast inequality machine. That’s certainly the case in those places that are now the main centers of quantum development, such as the US and China. In the US, for example, corporate and securities law and remuneration arrangements in combination with tax laws ensure that financial benefits from recent advances have flowed overwhelmingly to a very few. Mandatory arbitration and non-disclosure agreements favored by technology companies put pay disputes and discrimination claims. And they’re like out of collective sight and bargaining. This has become the model that other states typically aspire to reproduce, every state wants its own Silicon Valley, including Australia. I think this is something that we need to bear in mind when trying to envisage the regulation of quantum technologies or perhaps even more ambitiously, trying to imagine a quantum jurisprudence, nowhere can this undertaking be divorced from the inequality that is baked into our current global settings for financing and nurturing technological innovation.”

Michael J Biercuk: “I’ve been obviously very excited about developments in the field and it’s been an amazing transition to see research carried out almost exclusively in university laboratories, moving over not just to large corporates but also to venture capital backed startups like my own, like Q-CTRL. There is a lot of excitement about quantum that is not always well-informed, to the level that we have quantum dishwasher detergent. It has absolutely nothing to do with quantum, it’s just a sexy term. And the sexier something becomes the more likely it is to attract the attention of people who perhaps have money. This is something that is sometimes called ‘dumb investors,’ ‘dumb money.’ It’s understood in venture capital that if investors who are well capitalized, but not savvy about either technology or building businesses, start pouring money into the industry they run the risk of destabilising it because they can set terms for new companies that make them fundamentally un-investable in the future. My concern is that with all the excitement and attention that comes from legitimate successes in our community that we start attracting people who are not savvy investors, who may fundamentally destabilise our community, but may also lead to some kind of crash in the availability of funding for the more legitimate players. It’s a concern. I’m not saying that it’s all going to come crashing down, but it may get more challenging at times when we’re not delivering on the timelines that non-savvy investors would expect.”

Helen Nissenbaum: “And there have been tons of high-level principles, particularly from the AI industry, from the industry side of digital technology. And here are some of the examples. You see Microsoft and Google, of course, constantly trying to present themselves as responsible actors in this domain. Fairness, inclusiveness, transparency, as Gavin talked a little bit about, privacy, accountability, make the technology safe, incorporate privacy by design, and so on. And here are a few more of our favourites: Salesforce, talking about human value, alignment, transparency, safety. Facebook talking about responsible and thoughtful use of AI and concerned about the unintended biases that can happen when you apply machine learning or facial recognition and so forth, all of which they go ahead and do. Finally, here’s one from the European Commission. Similar things: safety, privacy, transparency, non-discrimination and so forth. And here’s the first one that you may be familiar with: Don’t be evil. Maybe we sum up all the ethical principles just with that. Don’t be evil. As if presenting these high level principles is going to yield more ethical behaviour. Who could argue with those high-level principles? And that’s what I want to talk about, because I think that unless we drill down into some of the more boring details, I mean, I don’t want to say boring, more complete. I’m not sure we’re accomplishing very much with those details.”

Kate Crawford: “The other concern for me here is, of course, the concern that persists with large scale A.I., which is the concentration of power that goes alongside who can actually build quantum computers. We’re seeing this already. Of course, it is many of the large companies that we know from the moment are again capturing Quantum. So, we have to think very closely around who will benefit most from these technologies and who stands to actually experience the greatest harms.”

Nikita Chiu: “I think the emergence of quantum technology has countless significant consequences for the whole international community, for all states. But I think it would be truly revolutionary if instead of conceptualizing development in this area as mirroring the space race or the arms race during the Cold War, but instead take this opportunity to use science and technology as a vehicle, a driver for international cooperation. I think that’s very much a possibility, as long as we start to conceptualize quantum as an opportunity instead of as a competition.”

Nikita Chiu: “I think instead of looking at different emerging technologies in isolation we should try to understand technology as a whole, because you can only build so many international regimes, like space regimes or nuclear regimes. We really need to rewire our thinking. Any possibility of using technology for malicious purposes is in our mind. If we do not try to use it for malicious purposes then technology would not pose any catastrophic consequences to society. So what I’m trying to do is very much to encourage alternative ways to think about emerging technologies that have a lot of uncertainty associated with them, because I think if we want to use technology to promote cooperation instead of competition, to improve our lives, instead of for malicious purpose then we can achieve that.”

Helen Nissenbaum: “If we accept this idea that ethics can be found in technology, then it’s possible to take this practical turn and this means looking at the creators and the regulations of technology and saying it’s not enough to take the technology is given, and then think about the ethical implications afterwards. One needs to think about the ethical issues from the ground up. Now, don’t be big and grand questions, “why technology favours tyranny”. But I think of the small, granular and gritty details that can be trivial details, that are settled by the technologist that can actually make the difference between promoting a certain ethical value or not. The values at play framework has these components which is discovering the values and implementing them in systems and then verifying. And I’m not going to spend any more time on the whole ideas, but I wanted to talk about the definition, why it’s so important to define these values in operational terms. Not just ‘do no evil’, but what does it actually mean in the context of a particular system?

So, I’m hungry. I haven’t eaten dinner yet. And this comes to me. We ask the question of dividing a pie fairly among four people. And I want to ask you how you do it. And if we were together, maybe you’d raise your hand and give me some ideas. But I suspect that one of the first things you might say is that you are divided equally among four people in this way, four equal slices. And then maybe someone in the room would say, but hang on a sec. That slice on the bottom right hand corner, if it’s right to you, is smaller. That was a bad cutting job. But then other people might object, and they may say, no, that’s not the right principle of fairness. Maybe the fair way to divide the pie is to give more to the people who need it. Maybe they’re hungrier and they need it. Or those who deserve it. Did you just run a marathon? Here, you can have the biggest piece. Or maybe those who are rich and they can pay more, or perhaps the winner takes all. So what I’m saying is that it really matters what concept of fairness is driving the particular system, design or application. It’s not just a technical question, what are we optimizing for? And it’s not simple with generalities. The same goes with privacy. And I’m not going to go into this case because I’m already running out of time. But when we come back to comments or declarations like Google’s ‘don’t be evil’, we have to ask, what does that mean in operational terms? And when we have Facebook saying, oh, we’re going to be responsible with AI, we have to say, well, how does that match up with these ideas? The fact that Facebook’s algorithms are known to exploit the human brain attraction to divisiveness, that they are known to be designed to polarize. How do we comport these high-level declarations with the gritty details?”