As the first blog entry, it seems appropriate to give some background about how The MetaMakers Institute came into being. Here goes…


I’m British – I like moaning!

And I’ve been grumbling for years about two things:

(i) that in University-led research, we hire the brightest and best researchers who do amazing work, yet have to leave the group after three years because of short term contracts, unless we are successful with further funding bids. Sometimes the work we do helps businesses to be more profitable and people in the company do well, yet the researchers who have helped with this still face unemployment at the end of their contract. This is not a great situation.

(ii) Computational Creativity research has so much potential, but has not yet had a real impact on cultural life. Software being autonomously creative, and acting as creative collaborators with people is so obviously going to be a big part of our life in the near future, but the large (and small) tech companies have yet to realise the potential for this. So, the only interesting independently creative software out there has been built as a prototype to test out some ideas in an academic research setting. This usually means that its not ready for public consumption, and getting it ready doesn’t lead to publications, so we tend not to do that.

In mid 2014, along with members of the Computational Creativity Research Group at Goldsmiths, most notably Michael Cook, I started discussing the commercialisation of our work. In our particular field, we build software platforms to test out theoretical ideas about the creative capabilities of software. Sometimes these platforms get quite sophisticated in terms of user interaction, indeed sometimes they have to be sophisticated to simulate software built professionally for particular experiments (i.e., sometimes it needs to look and feel like the real thing if we are going to test how people interact with the real thing). These platforms therefore represent a commercial opportunity: we could potentially sell the software directly, or sell time with the software acting as an automated consultant.

There are dozens of fantastic AI researchers worldwide who have built software that produces artefacts of real value (poems, paintings, theorems, musical compositions, fictional ideas, architectural and other designs, scientific discoveries, recipes, you name it …) in engaging, interesting and creative ways. And there are a few systems out there that perform live in interesting and creative ways, and some which contribute to cultural life through tweeting, etc. But it’s very difficult to get hold of and use these software prototypes, and (to the best of our knowledge) none of them are ready yet for the general public to enjoy.

After discussing the potential of these software platforms for commercialisation, the idea of The MetaMakers Institute was born. In the short term, we will undertake translational research to bring our own ideas and prototypes up to commercial quality, specifically by building Android and iOs applications for hand-held device. The plan is to commercialise these platforms and put the money into the research bank, so that when people’s contracts come to an end, there is a war-chest of cash which we can use to employ people. Of course, if the commercialisation is very successful, then we can use it to pump-prime new translational and even blue-sky research projects.

In the long term, The MetaMakers Institute will become a global organisation where AI researchers can come to spend time turning their research ideas into software that will change people’s lives: the idea is that the money we raise through commercialisation can be used to fund short-term visits for this purpose. Together, we can drag the world into a future where software creates for us, alongside us, and in order to challenge us.

The difficulty, of course, was finding the resources to undertake the kind of translational research required to get The MetaMakers Institute off the ground. Essentially, we were in a classic chicken and egg scenario, whereby we needed funding to be able to eventually raise our own funds. Then along came an amazing opportunity. The ERA Chair project – called Games Research Opportunities (GRO) – was won by Falmouth University and needed someone to run the project as the appointed ERA Chair. I pitched the idea of The MetaMakers Institute in the interview and in subsequent negotiations, and it became clear that the plan for commercialisation and impact fits the remit of the ERA programme very well.

So, here we are. It’s been somewhat of a slow start, as I had to move from Paris, I’ve recently become a father, and it’s taken a while to hire the new team (in fact, we’re still hiring …) But The MetaMakers Institute is very much up and running, and we are very happily housed in the Games Academy at Falmouth. We plan to get our first app out there in the next few months. It’s called Cillr, and it enables people to make entire games right there on their iPhone. We have to simultaneously solve the issues of how to design a user interface which enables all aspects of games to be altered using only your thumb; how automation can help you build better games more easily; and how to best commercialise the software. As a research platform, we will explore how we can get secondary AI systems to create games using Cillr, as part of our investigations into how to get software to create full games. We plan to build an API so that other people can write automated creators to build games with Cillr.

Making the MetaMakers