Making academia great again (from the business perspective)
How academia and companies can invent and innovate together
Academia is losing in its role of innovation leader as companies started investing heavily into fundamental research. It was not the case a decade ago (except pharma and few other domains) as fundamental research is costly for companies. Now, especially in AI, companies invest billions to win the race. It is hard to predict if funding large scale fundamental research in Deepmind, Google Brain or OpenAI with almost no direct business impact is sustainable in the long term, but let’s discuss its consequences for academia.
Draining brains from academia
Brain drain from academia to industry is normal when it comes to students. Educating talent for companies is the most important role of universities.
However, there is a brain drain also among senior level researchers and professors. This brain drain has been on the rise for a couple of years already, it is not healthy for universities and it is worst in trending domains such as Artificial intelligence and Data Science.
The positive aspect of the AI race is that it is very open and publications, data sets and source codes made by researchers in companies are often made available to everybody including research teams in academia (which is not the case in pharma and other fields). However, researchers in academia often lack access to live users so they cannot train machine learning models online on millions of users. They do not have proper computational infrastructures and engineering resources available to optimize and scale their prototypes. They cannot invest huge money to data labeling and they do not have enough funds to attract top interns.
There are attempts to halt the brain drain, but they are doomed to fail unless academia does not transform into an innovation partner for companies.
Academia is now perceived as a source of talent and impact of university research to business is underestimated not only by companies but also by experts and governments.
Building an ecosystem focusing on inventing, innovating and incubating in AI
In Prague, we are creating the role model of a sustainable innovation ecosystem that is based on balance among academia, corporate and startup worlds. I have co-founded prg.ai not only to educate more AI talent, but most importantly to make significant positive impact to business and society through AI innovations.
Our goals (programs) are aiming at
- building and cultivating the AI innovation ecosystem,
- educating and attracting AI experts and
- supporting excellent fundamental and applied AI research generating AI innovations and AI powered products for global market
at universities, research institutes, companies and startups together with Prague municipality and other partners.
In order to make this model work, we need (among other) to support academia in its role as innovation driver for companies, municipalities and other institutions.
Academia as innovation partner for companies
In prg.ai, we are working on guidelines for participants of the AI ecosystem. In this post, we will discuss how universities and research institutions should transform into innovation drivers for companies. Along other faculties, we are piloting this transformation at FIT CTU, where we already run an innovative program to streamline collaboration with companies.
The aim of the transformation is to increase the volume and quality of research collaboration with industrial partners along with ability to produce scalable research outcomes available for licencing or spinning off these outcomes into startups.
In the same structure as prg.ai programs, we first target structural changes of the ecosystem, then build capacities for research and ensure we have impact in all three components of the ecosystem — academia, corporations and startups.
The aim is to transform the faculty from supplier of excellent students to a role model of innovation driver helping corporations to transform while giving birth to new innovative companies.
First component is the ecosystem consisting of internal research capacities and external partners (companies, startups). Typically, companies are not interested to innovate with academia due to various reasons. They are not innovating at all, they cannot find relevant research capacities at academia or they prefer to innovate internally — hiring students and experts from academia.
It is not sustainable for companies nor universities pursuing (AI) research and innovations independently due to scarcity of top talents/experts, cost of high-end research and access to data, infrastructures and users.
Academia should work with partner companies to create ecosystem that enables sharing top talent/experts and their results. It is also important to build more capacities both in academia and in companies with invention and innovation capabilities.
Building, maintaining and improving the ecosystem
At FIT CTU we build and cultivate the ecosystem of 40+ collaborating companies. We are motivating companies into long term strategic research partnerships.
We are building resources at the university capable of delivering research POC (proof-of-concept) leading to innovations. Our focus is also fundamental research with potential of generating inventions that are useful for our partners. Majority of our projects including contract research are negotiated under non-exclusive licencing scheme allowing university to build on top of previous results and licence technology to more non-competing partners. Successful collaborations are typically supported by joint laboratories where university and company experts collaborate with students on inventions and innovations. Below is the agenda needed to handle relationships with companies if you like to be perceived as innovation driver.
Ecosystem bootstrap and maintenance (BUILD)
First we need a set of simple and transparent rules that will help participants to navigate in complex environment of academia — business relation and transfer of academic IPs.
Setting up and support of commercialization processes
- internal rules how to approach IP transfer, setup spinoff company and related
- partnership agreements with companies, templates for contract and licencing agreements,
Then somebody has to take care of existing partnerships and push them forward. Typically you have to separate administrative part and expert part.
Maintaining and developing partnerships with companies
- agreements extension, invoices and other administrative tasks
- expert work with existing partners to prepare topics of projects and long term research agenda
- reporting status of services provided by university to partner companies, reporting results of collaboration to academia
For reporting, you need detailed numbers and information about related activities.
Ecosystem evidence, collecting and measuring results
- licences, patents, contract research projects, collaborative research funded by third party agencies
- internal experts, IP for licencing, opportunities for investors and donors
- evidence of business partners and collaborating companies
- evidence of possible investors and donors
Connecting to right business partners (CONNECT)
It is always good to actively seek for companies that would benefit from your expert knowledge and technologies for licencing. It is better to collaborate with companies that are willing to support and apply research with publication potential. To strengthen partnerships, you need to organize regular matchmaking events for local companies seeking talents and possibly in need of some of your expertize or technology.
- Seeking partners with innovation potential
- Seeking investors for scalable technologies and startups
- Seeking donors for open source projects, non-profit research
- Connecting with other research institutions for joint collaboration with companies
- Connecting and collaboration with NGOs, public institutions, municipalities (public responsibility of the university)
- Matchmaking events for companies, alumni, experts, students (such as our CO@FIT)
- Software support for assigning contract projects to students, research projects to experts, possible research topics for collaboration, matchmaking experts from academia and industry
Promotion of research and transfer capabilities (PROMOTE)
Dissemination of success stories and examples of academia-business collaboration outcomes is the best way to motivate others. It is also about improving the environment, explaining rules and communicating your values. You need to target multiple stakeholders — professors, students, company experts, their HR staff and external audience to bring more talent in.
- Web presentation and marketing materials for partners, students and staff
- Promotion of your partner companies to students and staff (including matchmaking)
- Promotion of your internal capacities outcomes, references and success stories (writing and promoting content based on data collected in BUILD).
- Dissemination of your activities and collaborations.
Values
While building the ecosystem, it is important to set and respect values that all participants should follow. Academia is largely funded from public money and you have to ensure this money is not spent in a way that can create an unfair advantage for a particular private company. It is also important to monitor, report and respect conflicts of interests as people involved are typically active in multiple businesses (and I am no exception). The ecosystem should protect the interests of academia, but also of commercial partners and students. Goal is to inspire and bring opportunities not to force anyone into collaboration. Openness, respect, responsibility and honesty will help you to create a sustainable environment.
Capacity building (PEOPLE)
Most of academic experts have their goals and KPIs aligned with educational and fundamental research activities. They feel distracted when you ask them to deliver a contract research project for a company. Of course it depends on the attractiveness of the topic and how far the project is from fundamental research that can be published. You should work with your staff, educate and motivate them towards inventions and innovations for partner companies. You should educate students in this direction and attract experts from outside. Moreover you need to help partner companies in their internal capacity building process so they can accommodate your inventions and innovations. Here are the steps we take.
Educating students and staff towards delivery for companies and commercialization (TEACH)
At FIT CTU, we have a large number of talented students. At the same time, there is a tremendous amount of job opportunities in Prague region and companies are hunting even students who just enrolled. Our aim is to increase the number of students capable of inventing and innovating in our labs or in our partner companies. Also we train students and staff towards ability to deliver and support their entrepreneur spirit.
- Training students towards research in general (see our research summer)
- Encouraging students to join research projects and supervise them
- Educating students and staff to improve in project management, executive roles, entrepreneurship and budgeting
- Training staff in presentation of their research to business partners and investors
- Helping partner companies in capacity building and staff education towards strong innovation culture
- Building internal teams focusing on research agenda in strategic areas with high commercialization potential
Attracting talents and senior experts (COME)
People all around the world are looking for opportunities to grow and work in a great environment. Even when they learn about your ecosystem, you need to be prepared to accept them and streamline the process to resettle. Actively seeking experts in areas of your interest respecting research needs of your partner companies is important for building relevant capacities.
- Adjustment of environment enabling smooth on-boarding and work of foreign experts in your laboratories
- Industrial PhD — working with your partners to prepare research topics and funding
- Attracting foreign students with research potential and integrating them into projects
Retention and systematic development of research capacities (STAY)
It is always easier and cheaper to prevent churn of your staff than hiring their replacements. In academia, it is beneficial when experts migrate, learn in different environments and visit top global labs. You need to make sure they are motivated to come back. Also leaving academia to industry is one of the biggest streams of academic churn as discussed in the introduction. Making academia competitive to industrial research positions is the only way how to retain top talents.
- Follow-up funding for students who demonstrated research potential and are about to graduate.
- Securing extra funding for doctoral students and their fundamental research (donations, grants).
- Actively seeking for follow up projects for staff employed on projects and contracts that are about to end.
Diversity
We are encouraging diversity in the academic labs so people of different gender, origin, culture or background feel welcomed. People should be treated equally and get opportunities based on their results. When evaluating results we rely exclusively on data indicating quality of results and satisfaction of our clients (partner companies).
Impact to business and society (IMPACT)
Having positive impact to business and society is increasingly important role of academia. Your research should be relevant and tackle problems and challenges recognized by partner companies and society as important. Of course you can also find a niche and your research outcomes can give birth to a successful company, but is always good to question the relevance and potential of the research done in your laboratories. As a technical university with a long tradition in collaboration with industry, we feel responsibility for competitiveness of our society and companies employing people in the region. We help corporations with the AI transformation process, company innovations and we also invest into research and development that can generate new scalable business opportunities. The goal is to develop solutions to problems seen in many companies so we can licence the same technology multiple times. Such research outcomes bring not only higher efficiency and impact, but also higher revenue from licencing.
Supporting fundamental research leading to applications (INVENT)
Fundamental research is hard and costly. However, when you have potential applications in mind, it can bring competitive advantage that is way bigger and harder to copy than anything that can ever be obtained by applied research. At universities, fundamental research is often funded by public grant agencies. As companies compete in self driving or conversational intelligence technologies, they are increasingly keen to finance fundamental research as the only way forward. Academia should get better in delivering contract fundamental research to companies.
- Mapping suitable directions of fundamental research that will pay-off your investment (staff, internal funding).
- Assisting your researchers to get matching funds from public sources or from companies (to make fundamental research affordable).
- Helping companies to understand the importance of fundamental research enabling disruptive innovations.
- Investment into IP protection (patenting, contracts, licencing)
Supporting applied research for companies (INNOVATE)
Contract research for companies is becoming an important source of university funding and should make up a significant percentage of total budget for technical universities.
- Identifying INVENT outcomes (and outcomes of third party cutting edge fundamental research in general) with an innovation potential for partner companies
- Assisting your researchers to get matching funds for applied research, look for non-competing companies that can share results.
- Helping companies to identify innovation directions (e.g. AI transformation roadmap)
- Actively searching for companies with an innovation potential and negotiating contracts
Cultivating scalable research outcomes and startups (INCUBATE)
The most precious outcomes of your research are those with a potential to scale into successful global product. Top universities implemented systematic support for university spinoffs decades ago, however in some regions (including Czech Republic) founding a university spinoff is almost impossible. There is a governmental initiative aiming to change it, but it will not succeed unless academia change the mindset and internal processes that reduce the likelihood of larger scale academic churn to startups (we experienced some examples already). Also preseed and angel funding is largely not available in spite of the fact that it can cost as little as $50 thousand to develop an MVP of AI product at academia in Prague (what a contrast to the US).
- Searching for funding opportunities of scalable innovations (investors, companies)
- Looking for customers willing to licence existing technology owned by academia
- Investing your own resources and funds to spinoffs in exchange of a profit share
- Managing the portfolio of scalable IP, licences and shares
Humanity
We make sure that our research, innovations and startups have a positive impact on society. We avoid research that directly threatens human rights or interfere impermissibly with human privacy. When licensing, we make sure that our technologies do not fall into the hands of authoritarian regimes or serve military purposes.
Making academia and companies innovate together, at scale
As you see above, turning academia into an innovation partner for companies is a lot of work. I have spent the last 15 years at academia trying to improve things in this direction. With a very limited resources we managed to bootstrap income from collaborations and also developed software tools to support the agenda.
It is however very hard to finance larger scale software development projects at a single faculty. Companies are also willing to use one software to collaborate with all academic partners. And software should be accompanied by services and senior consultants to match expectations of corporates. The lack of single/simple interface often prevent them to invest time of their experts and they better innovate internally.
We also wanted to have a global impact as we recognized that the above mentioned problems are widespread. Therefore we founded unico.ai company back in 2017 and raised seed money to enable academia and companies innovate together at scale.
We have just releasing a software platform experts.ai enabling companies to find experts or investment opportunities in academia and facilitate their collaboration. So far, we started with experts in the Czech Republic, we are adding Poland and Slovakia by the end of the year.
Our software and services are already helping academic institutions to become better innovation partners for companies. At the same time, we convince companies to invest at academia so they can boost their innovation capabilities. Win-win collaboration model is essential for further challenges as talent is scarce and shared resources at academia can help you to innovate cost-efficiently and benefit from disruptive inventions when funding fundamental research.
Conflict of interest disclaimer: When writing this article, I was switching my roles as
- AI associate prof. at FIT CTU in Prague,
- co-founder of non-profit organization prg.ai,
- vice-dean for collaboration with industry at FIT CTU in Prague,
- co-founder of the unico.ai company.
My feeling is that in all roles my aim is to make academia better in delivering innovations to industry at 1) department level, 2) meta-university level, 3) faculty level and 4) global scale. I believe that the interests of my organizations in this sense are rather corresponding than conflicting. I always do my best in protecting the interests of all organizations I represent. Of course there always will be a bias and I apologize to other prg.ai faculties and possible unico.ai competitors for not giving them equal visibility in the article.
