- Effective risk management (paradigm shifts to models, regulatory, exogenous factors)

- Regulation: Who is responsible for ensuring good practice in the use of data?

- Reacting and coping with market regime change and extreme events eg Flash Crashes, election outcomes, unpredictable news

- Age and depth of reliable data: are models factoring in unusual/unseen market conditions?

- Embracing and processing quant concepts/cultures in a fundamental house

- The blended approach: using fundamental research with algorithmic optimisers

- Using an optimiser to suggest trade ideas and comparing to fundamental research

- Concerns about the future of fundamentals in a quant world

- Purely systematic trading: data scientists/trading team watch, analyse and execute

- Interpretability issues: where is the regulation and the trust in model selected statistics

- Over defined models that are led to give an output

- When a model can’t cope with a market regime change

- Coping with sudden extreme exogenous factors: is your model prepared?

- Sourcing the newer and fresher forms of alternative data

- Coping with a labour intensive cleaning process

- Signal mining: Finding the balance between big insights in unique/exclusive data or finding weaker hidden  levels of insight from widely available data

- Encouraging suppliers to follow standards and invest in the opportunities of selling data

- Monitoring the direction of Alt Data: where will privacy/compliance go?

- Applications of different types of AI (NLP and language barriers/translation, machine learning, genetic algorithms)

- Can your system truly teach itself new skills or is it operating in the method and world you define?

- Uses in various stages of investment process (eg cleaning/labelling, optimising trade ideas, executing)

- Signal hunting: Working in blackbox style with own statistics (interpretability above)

- Open source concepts: working with a community to share ideas and concepts to help towards the next big breakthrough

- Is it truly AI that can regenerate, self learn and adapt or is it a defined algorithm working in a set universe that is open to tail risk and market regime change?

- Going back to the proven fundamentals: AI as a science

- AI advance: mini breakthroughs and where to find them (the biggest impact is processing power)

- How can you quantify a non-standard universe

- What is the impact of quantum computing? How far is it really and will you be able to get new systems in whilst there is still trust in the safety of capital value?

- Blockchain: the benefits and the fallacies

- Data usage platforms 

- High Performance Computing

- Speed of processing data

- In house systems vs vendor built