Women in AI: Catherine Breslin helps companies develop AI strategies

To give women AI academics and others their well-deserved – and overdue – time in the highlight, TechCrunch is publishing a series of interviews specializing in the extraordinary women who’re contributing to the AI ​​revolution. We’re publishing these articles all year long because the AI ​​boom continues, highlighting key work that usually goes unnoticed. Read more profiles here.

Catherine Breslin is the founder and director Kingfisher Laboratories, where he helps companies develop AI strategies. She has spent over 20 years as a man-made intelligence scientist and has worked for the University of Cambridge, Toshiba Research and even Amazon Alexa. Previously, she was an advisor to VC fund Deeptech Labs and director of solutions architect at Cobalt Speech & Language.

She accomplished her bachelor’s degree on the University of Oxford after which obtained her master’s and doctorate degrees on the University of Cambridge.

Briefly speaking, how did you start in artificial intelligence? What drew you to the sphere?

I at all times liked maths and physics in school, so I made a decision to review engineering at university. This is where I first learned about artificial intelligence, even though it wasn’t called artificial intelligence back then. I used to be intrigued by the thought of ​​using computers to process speech and language, which is straightforward for us humans. I then began a PhD in voice technology and worked as a researcher. We’re at some extent where there’s been lots of recent progress in artificial intelligence, and I feel like there is a huge opportunity to create technology that may improve people’s lives.

What work in AI are you most happy with?

In 2020, at first of the pandemic, I founded my very own consulting firm with a mission to offer real-world AI expertise and leadership to organizations. I’m happy with the work I’ve done with my clients on different and interesting projects, and that I’ve been in a position to do it in a extremely flexible way around my family.

How do you take care of the challenges of the male-dominated tech industry, and by extension, the male-dominated AI industry?

It’s hard to measure exactly, but about 20% of the AI ​​field is made up of ladies. In my opinion, this percentage decreases with age. For me, among the finest ways to take care of that is to construct a support network. Of course, support can come from people of any gender. But sometimes talking to women who’ve been in an analogous situation or seen the identical problems is reassuring and it’s great to not feel alone.

Another thing for me is to consider carefully about what to spend my energy on. I imagine we are going to only see lasting change when more women take up senior and management positions, and that will not occur if women devote all their energy to fixing the system slightly than advancing their careers. There must be a realistic balance between pushing for change and specializing in your personal day-to-day work.

What advice would you give to women trying to enter the sphere of artificial intelligence?

Artificial intelligence is a big and exciting field with lots happening. There can also be an enormous amount of noise surrounding what can look like a relentless stream of documents, products and models being released. It’s unattainable to maintain up with every thing. Moreover, not every article or study result shall be significant in the long term, regardless of how flashy the press release. My advice is: find a distinct segment that you just actually need to grow in, learn every thing you possibly can about it, and tackle the issues that you just are motivated to unravel. This offers you the solid foundation you wish.

What are probably the most pressing issues facing artificial intelligence because it evolves?

Progress over the past 15 years has been rapid, and we have seen AI move out of the lab and into products without having to step back to properly assess the situation and anticipate the implications. One example that involves mind is how much of our voice and language technology works higher in English than in other languages. This doesn’t mean that researchers have ignored other languages. Significant effort has gone into the technology of languages ​​aside from English. However, the unintended consequences of higher English technology mean that we create and implement technology that doesn’t serve everyone equally.

What issues should AI users pay attention to?

I believe people ought to be aware that AI just isn’t a silver bullet that may solve all problems in the following few years. Creating a formidable demo could also be quick, but it surely takes lots of effort to construct an AI system that performs consistently well. We shouldn’t lose sight of the undeniable fact that artificial intelligence is designed and built by people, for people.

What is one of the best approach to construct AI responsibly?

Building AI responsibly means incorporating diverse views from the very starting, including those of shoppers and anyone impacted by your product. Testing your systems thoroughly is vital to make sure you understand how well they perform in various scenarios. Testing gets a status for being boring in comparison with the joy of coming up with recent algorithms. However, it is amazingly necessary to know in case your product really works. Then it is advisable to be honest with yourself and your customers about each the capabilities and limitations of what you are constructing in order that your system doesn’t get abused.

How can investors higher promote responsible AI?

Startups are creating many recent applications of artificial intelligence, and investors have a responsibility to rigorously consider their financing decisions. I’d like to see more investors speak out about their vision of the long run we’re constructing and the way responsible artificial intelligence suits into it.