AI is driving rapid change in the tech sector. New advancements in the field have brought tools like ChatGPT, Bard, and StableDiffusion to the public, while also offering the potential to solve knotty problems in areas such as medical research and climate change.
But from the designing phase, right through to training and implementation, AI models must be made with ethical boundaries in mind. Without careful consideration of input and end result, AI can reflect the prejudices or limitations of its developers, and cause real harm.
In this episode, Rory and Jane speak to Dr. Sohrob Kazerounian, distinguished AI researcher at cyber security company Vectra AI, to unpack the considerations that go into making AI ethical, and how firms can work to eliminate unexpected bias.
Highlights
“Whereas for the last 5-10 years, we've been telling people in security, ‘hey, you need AI, you need machine learning’, we're now starting to recognize that machine learning people need to start thinking about the security and robustness of their models in a way that might be different to what they're used to.
“I think it becomes incumbent on people who want to use machine learning and AI techniques to really spend the effort to do it. Whether or not in the limit we can eliminate all bias is a sort of open philosophical question and debate we can have. But I think in the near term, it's going to be something that we definitely have to at least attack and try and try and eliminate”
“There's no world in which you can basically let these models just put a large number of people out of their jobs, without also having to say ‘has it benefited from their labour?’”
Footnotes
- What is ethical AI?
- What are the pros and cons of AI?
- Workday hit with claims its AI hiring systems are discriminatory
- Can Microsoft's new approach to AI erase the memory of Tay?
- OpenAI tool previously thought 'too dangerous' for the public goes generally available
- Is the UK falling behind the EU on AI regulation?
- EU's AI legislation aims to protect businesses from IP theft
- Why are AI innovators pushing so hard for regulation?
Subscribe
- Subscribe to The ITPro Podcast on Apple Podcasts
- Subscribe to The ITPro Podcast on Google Podcasts
- Subscribe to The ITPro Podcast on Spotify
- Subscribe to the ITPro newsletter
- Subscribe to ITPro 20/20
-
Bigger salaries, more burnout: Is the CISO role in crisis?In-depth CISOs are more stressed than ever before – but why is this and what can be done?
-
Cheap cyber crime kits can be bought on the dark web for less than $25News Research from NordVPN shows phishing kits are now widely available on the dark web and via messaging apps like Telegram, and are often selling for less than $25.
-
Google Cloud Next 2025: Targeting easy AIITPro Podcast Throughout its annual event, Google Cloud has emphasized the importance of simple AI adoption for enterprises and flexibility across deployment
-
Is the IT service desk dying out?ITPro Podcast Employees could soon use chatbots to resolve common IT issues rather than wait for tickets to be resolved
-
Harnessing unstructured data to fuel AISponsored Podcast Knowing how your data is stored can make all the difference to your AI success
-
How effective are AI agents?ITPro Podcast AI agents have been sold with the promise of fully automating tasks without the need for step-by-step instructions
-
Are most AI projects destined to fail?ITPro Podcast The current era of AI projects could provide the lessons we need to make the most of the technology going forward
-
Revenge quitting and Microsoft goes quantumITPro Podcast Breakthroughs in quantum computing across February have been matched by financial and workforce upsets in the tech sector
-
Who wins out from DeepSeek's success?ITPro Podcast Firms focused on AI at the edge and open source LLMs could reap massive rewards from the Chinese startup's industry upset
-
DeepSeek flips the scriptITPro Podcast The Chinese startup's efficiency gains could undermine compute demands from the biggest names in tech
