Security
week or 10 hours a week by implementing the solution. Think about how AI is going to offer the business advantage and consider questions like: What is the Return on Investment (ROI) or long term value? When presenting to a board of directors, it’s crucial to express it in their terms to understand the impact to the organization, or how is it going to help the company perform better than the competition? Another consideration is what are the risks if we don’t adopt AI?
Start small, scale smart If the organisation is not using AI extensively, it’s best to begin with a pilot project to demonstrate AI’s value before scaling up. It might initially be an AI proxy for filtering work- related content. So, users can use things like ChatGPT, but the security teams can filter the content that is uploaded to it to ensure no sensitive data or intellectual property is exposed as data loss prevention is critical to an organisation strategy or position. This strategy allows for a competitive edge by allowing users to use generative AI but with the ability to protect sensitive information or IP overall.
Monitoring AI in action For practitioners, there are some initial steps to consider when using AI which will help in the long run as they work up to communicating the business’ use of AI to the boardroom. Firstly, it’s important to identify what is the problem that they’re trying to solve and, if any, how AI can improve efficiency gains, decision-making, or mitigating bias? Understanding these goals helps define the risks that need to be monitored, such as bias in algorithms or unintended consequences. From there, performance tracking metrics or KPIs can be set to align with these goals, including production output (if efficiency is the goal), customer satisfaction scores, or even the accuracy of AI predictions. It will also help to have an idea of how transparent the AI systems are, which means understanding how they arrive at decisions and being able to explain them in clear, non-technical terms. Data governance must also be a priority in order to establish
clear guidelines on data collection, storage, and usage. Each industry and geographic location will have specific data privacy and compliance regulations that will need to be adhered to. Be prepared to communicate to the Board how these regulations are being met, or indeed where they present potential areas for improvement. Finally, maintain human oversight of all AI systems - without letting them run wild (or unmonitored) to ensure responsible use and to intervene when appropriate. Don’t forget that monitoring AI is an ongoing process. As
the organisation’s AI use evolves, so should its monitoring practices. Regularly review and update these strategies to maintain continued responsible and effective AI implementation. There are frameworks to assist in this endeavour; for example, Google’s secure AI framework (SAIF) that takes into account six core elements for a safer AI ecosystem that are designed to help ensure that when AI models are implemented, they are secure-by-default.
www.pcr-online.biz
EU AI Act In terms of regulations the boardroom must also consider that the recently introduced EU AI Act shows the EU’s commitment to AI governance. It balances fostering innovation and ensuring that AI serves society more for good than malicious activities. As the regulatory landscape evolves, policymakers must continuously engage and collaborate with AI industry leaders. It will ensure that they navigate the complexities of AI and harness its potential for positive engagement within society. The AI Act’s cybersecurity requirements emphasise the
need for robust security measures in AI systems. The Act ensures resiliency against cyber threats and potential misuse by mandating rigorous testing and documentation. It is particularly crucial given the increasing sophistication of cyberattacks targeting AI vulnerabilities. For cybersecurity professionals, the Act presents an
opportunity to focus the necessary resources on securing their AI data and infrastructure, promoting the development of AI systems that are not only innovative but also secure by design. Furthermore, the Act stresses the importance of transparency and accountability in line with foundational cybersecurity principles. Organisations must implement comprehensive GRC (governance, risk compliance) frameworks to ensure compliance, such as ISO 42001, NIST AI Risk Management framework, or other similar frameworks that are publicly available and implement security functions or processes at every stage of the AI lifecycle. This proactive approach to AI regulation can help mitigate risks, build brand trust, and pave the way for a secure digital world.
Invest in education Finally, it is important to ensure the organisation has a basic understanding of AI capabilities and limitations and extend this education right up to the Board level. Building an AI-aware workforce will create an AI-savvy culture, and no one will get left behind. Work internally with different departments to understand where AI can help and introduce it gradually, with consistent training on acceptable use, how it should be downloaded, and when. Invite discussions on AI. By demystifying AI and highlighting its strategic advantages,
directors can lead their companies towards a future fuelled by intelligent technology. Remember, AI is not a replacement for human intelligence but rather a powerful tool to augment it. It’s crucial to recognise that AI is not a replacement for human decision-making, but rather a collaborative tool, like a trusty sidekick. Just as a superhero relies on their sidekick for support and guidance, we must view AI as an augmentation of our abilities, not a substitute. The temptation to let AI make wholesale decisions on our behalf is like the siren’s call, luring us towards the rocks of unintended consequences. By embracing AI now, companies can seize the competitive edge and secure their long-term success. Te challenge of unlocking the full potential of AI while
mitigating its risks is one that we must rise to, for the sake of our future and the generations to come, like a relay race where we pass the baton of responsibility from one generation to the next.
July/August 2024 | 41
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52
