Demystifying Artificial Intelligence
Summarising some of the highlights from the recent Aerospace Xelerated panel on AI.
Aerospace Xelerated recently hosted a panel discussing one of our upcoming programme themes. In the ‘Demystifying AI’ panel, we welcomed some exciting guests from the field:
- Sitah Alotaibi: Information and Systems Analyst, and cofounder of OURZ
- Wasmiyah Alhayyan: Computer Teacher, and the other OURZ cofounder
- Gianluca Mauro: Cofounder of the AI Academy
- John McKenna: Founder and CEO of sees.ai
The diverse range of panellists, hosted by Programme Manager Ksenia Kurileva, discussed the importance and challenges that Artificial Intelligence poses. We learned how panelists got into the AI space and their experiences with artificial intelligence, including the projects that they are currently working on.
John McKenna shared how Artificial Intelligence is assisting in aviation regulation in relation to drone company Sees.ai. He flagged the issue of capturing data and its importance to AI: data is required to train AI models, but the initial data needs to be obtained before that can happen. McKenna discussed the need for a piece of technology that regulators are comfortable with, which uses deterministic technology and can be flown by a human pilot supported by a range of technologies.
Sita then talked about the need for data to personalise their parenting app, OURZ. The app will be using AI to connect the regular users with specialists. AI will analyse posts that parents publish in the app, using the information to connect the parents with the appropriate specialist who is best suited to respond.
Gianluca emphasised the need to look beyond the volume of data. He believes that there can be too much of focus on big data, when there are other important factors to consider, such as data accessibility. He stressed that accessibility is an important problem for companies to solve, particularly large organisations that may be slow in adapting to AI and/ or developing an AI culture. He also highlighted the importance of the quality of data and collecting the right data for training AI models. He raised the issue of data-centric AI, which focuses on the kinds of data that algorithms are fed with. Later, he expanded on what might make bad data, citing a company that he worked with that had many years of data that wasn’t very organised. This meant that a lot of time was required to make sense of the data and how it might be useful from an AI point of view.
Beyond data, the panel also discussed another essential element of AI: the people, and their talent. They discussed the need to have the right technical knowledge and experience in place, and the need to hire people from the right spaces.
Inclusivity in AI was also a key topic in the panel’s discussion. Ksenia suggested that although AI and techniques like natural language processing are used by startups, it often seems as if the knowledge is in the hands of data scientists or tech experts. This could lead to people in operations or sales and marketing feeling shut out of what they need to know. Gianluca agreed that this lack of inclusivity and understanding across whole company structures can be a big problem and acknowledged the need for company-wide cooperation when using AI to solve company challenges. He suggested that it is very important to break down walls and bring together those working in both the tech and business sides.
Talking about when to start charging for an AI-based product, John discussed the need to go beyond a minimum viable product to show customers that your product has the potential to radically change their world. He flagged the difficulty of balancing the need to release a product as soon as possible with ensuring you have a complete solution that actually solves your intended customer’s problems.
The panel also talked about the need to secure funding for projects and how to best align management buy-in too. The panel discussed pilot projects and the need to use “low hanging fruit” to get things started, get management on board, and spread the excitement about an AI-based project. Panellists touched on the need to get people who are skeptics or fearful of new technologies to accept that AI can provide new solutions and help everybody. Starting with a project that doesn’t take a long time to build helps to show others what AI can do and the potential of a larger project.
Helpful resources on the AI journey were another topic that was covered during the panel. The panelists talked about communities, network accelerators, and courses that could provide support for people growing startups with the use of AI. They stressed the need to continue learning and educating yourself, as well as finding opportunities to meet with investors and other important contacts.
During Q&A, the audience asked to what extend AI would replace humans. John emphasised that, while some tasks — particularly those that might be repetitive, dangerous, or undesirable — could be replaced by AI, automation and robotics, human talent and creativity are still required to make this happen. Gianluca pointed out that AI and robotics should be looked at as being able to replace single tasks, rather than whole jobs. By staying abreast of what’s happening in AI, many people can make sure they have the skills and knowledge to be part of this technological journey.
Have you developed a novel software solution in the AI and/or Autonomy theme that could be applicable to aerospace? To apply for this cohort, head over to our website.
Learn more about Aerospace Xelerated in our FAQ or watch the recap of our Ask Us Anything webinar. You can also book an Office Hours call to discuss your queries with the programme team.
Stay up-to-date with our latest updates via Twitter and LinkedIn and subscribe to our mailing list!
