At last week’s Ai and Big Data Expo, hosted at London Olympia, there was yet more evidence that 2023 will be looked upon as the year that Artificial Intelligence finally drove out of academia, labs, data-science spaces, and landed squarely in the mainstream.
I’ve chaired several of these large events for the Ai Expo, and this was by far the largest both in scale and number of attendees across both days. It was fantastic to see such a diverse range of speakers and attendees taking part in a wide range of topics and conversations. It was notable that this year appeared to have a lot more students in attendance - exploring what their careers might be and what jobs they might want to seek.
Speakers included Ebay, LNER, ITN, Google, MSQ alongside numerous other businesses navigating the Ai space.
Main Themes and Topics
The rapidly evolving fields of Generative Ai, Causal Ai, and Predictive Analytics dominated the sessions, alongside playback on how advanced Machine Learning is now firmly part of the daily running of most large businesses, to great effect.
On multiple occasions speakers and panelists discussed the complicated economic conditions the world is facing, and the extent to which this situation has been a huge driver of Machine Learning uptake - to reduce costs and increase profits.
Here is a summary of the other important topics that were raised throughout the event:
· Ethical Use and Bias Mitigation: Many speakers noted issues around the ethical use of Ai and the need to ensure Ai systems are fair and unbiased should be a major concern and top priority for anybody driving data-driven decision making. Ai can inadvertently perpetuate biases present in training data, leading to unfair outcomes, which has been reported for well over a decade now. Because of the new explosion in Ai use, via APIs and openly accessible architecture, we’re seeing a big increase in decisions being made laced with bias and errors. It's crucial for businesses to implement measures to detect and mitigate these biases, to maintain trust and avoid potential legal issues. These included regular audits, diversifying training data, and involving ethicists in Ai development. Companies should prioritise creating Ai that aligns with ethical standards and societal values, not just business objectives.
· Data Privacy and Security: With the increasing reliance on Ai, businesses who handle large amounts of data need to make data privacy and security paramount. Companies should be investing heavily in secure data storage, encrypted communications, and regular security audits before embarking on large-scale Ai productisation. This protects against data breaches, maintains customer trust, and avoids legal penalties.
· Integration with Existing Systems: A lot of speakers talked about the challenges of successfully integrating Ai into existing business systems as being a complex but essential task. Ensuring new Ai solutions work seamlessly with current infrastructure is often not possible because of the incompatibility of legacy systems. Businesses should conduct thorough analysis and planning, often involving IT specialists, to ensure compatibility and efficiency. This step is critical to maximise the benefits of Ai and avoid disruptions to ongoing operations. Predictive analytics in Ai might become an important tool for risk assessment in industries like Financial Services, FMCG, Healthcare and the Public Sector when assessing how well Ai will affect existing services and outcomes. Organisations should use automation to identify potential risks, assess their impact, and develop mitigation strategies. This proactive approach to risk management can save significant resources and protect against unforeseen events.
· Scalability and roll-out of Ai-based systems: Linked implicitly to the integration issue is the challenge of scaling up and rolling out new services. As businesses grow, their technology systems must also scale accordingly. Ai has a pivotal role to play in this, but also presents challenges. Scalability involves ensuring Ai solutions can handle increased data loads and more complex tasks without performance degradation. Businesses should design Ai systems with scalability in mind, using cloud services and modular, composable architectures. This ensures that Ai systems remain efficient and cost-effective as the company expands.
· Cost Benefit Analysis: This was mentioned repeatedly. The cost of successfully implementing Ai involves a lot of investment in development, training, and maintenance. 2023 saw a lot of businesses throwing Ai at the wall and hoping it would stick to a problem, but mature industries need to be conducting robust cost-benefit analysis to fully understand the financial implications and potential return on investment. This analysis should consider direct costs, potential revenue increases, efficiency improvements, and long-term savings. The message was loud and clear; Don’t do Ai or productisation if there is no solid business-case.
· The role of regulatory compliance and governance: Because Ai will increasingly be subjected to regulations governing data use, privacy and decision-making processes, it’s important that governance is baked into all the processes surrounding a businesses use of the tools. Implementing successful guardrails to test and assess outputs and outcomes from generative Ai models and LLM recommendations will be critical. Businesses must stay informed about relevant laws and ensure their Ai systems comply. This involves understanding the legal landscape, consulting legal experts, and regularly updating Ai knowledge amongst employees.
· Talent Acquisition and Training: The success of Ai initiatives depends heavily on having skilled professionals in place. This relates to both those using the tools to supplement their roles, as well as having focused Ai experts and data-science talent within organisations more broadly. The primary focus for businesses should be on recruiting talent with some Ai expertise, but also in providing ongoing training to their current workforce.
· A culture of Innovation and New Product Development: Culture was cited as being key to driving adoption of new Ai-led successes. But also discussed was the ability of Ai itself to drive innovation - by analysing trends, generating creative ideas and optimising product designs. Businesses should be using automation to identify market gaps, forecast trends, and develop products that meet emerging needs for themselves and for clients. Whilst this should be ingrained in all staff as part of their roles, having a dedicated, proactive, and empowered core team analysing and assessing opportunities will keep companies relevant and competitive.
· Sustainability and Ai: There were several talks throughout the event that touched on the role Ai will play in promoting sustainable business practices, by optimising resource use and reducing waste. Conversely, there was also discussion of how Generative Ai especially consumes vast quantities of data processing and is already leaving a huge carbon footprint. Companies should explore how Ai can contribute to their sustainability goals, such as via energy-efficient operations or sustainable supply chain management, but at the same time make sure they’re sourcing products and partnerships that meet their ESG goals. This not only benefits the environment but also resonates with increasingly eco-conscious consumers.
· Real-time Decision Making: I heard a lot about Causal Ai, which is a form of artificial intelligence focused on understanding and modeling cause-and-effect relationships, rather than just correlations. It goes beyond traditional predictive analytics by not just forecasting what will happen, but also explaining why it will happen. Giving businesses a deeper insight in this way allows them to make more informed decisions earlier, particularly in relation to strategic planning and policy development. Or in deciding how to invest in future product development or rewards for customer groups. Implementing causal Ai will lead to more effective interventions, as businesses better understand the underlying mechanisms driving outcomes. These enhanced decision-making processes can also lead to better resource allocation, targeted marketing strategies and improved product development. Of all the themes being discussed this appeared to be the one that might play the most disruptive but significant factor in all future Ai work.
Events like the Ai and Big Data Expo are important melting pots of ideas, voices and emerging trends. Whilst not exhaustive the themes discussed above seem to encapsulate the key areas for businesses and service providers to focus energy on. In 2024, as businesses continue to navigate a landscape increasingly shaped by advancements in Generative Ai, Causal AI and Predictive Analytics, it’s never been more important to embrace disruption, but also to prepare to mitigate the risks associated with them.
Partnerships and collaborations will facilitate Ai advancements, while businesses must also address its impact on employment and stay prepared for future developments. These topics are all pivotal in guiding how businesses use, build and deploy Ai in the coming year, ensuring they remain agile, informed, and competitive in a rapidly evolving digital landscape.
MMT offers expert consulting and cutting-edge technology implementation, ensuring your business harnesses Ai's full potential in ethics, integration, innovation and efficiency, guiding you confidently through the evolving digital landscape with tailored, future-proof solutions. Get in touch to find out more.