The tech community needs to change the way it views Generative AI.
That is according to Lenovo SVP and chief information officer, Arthur Hu who told the main stage audience at London Tech Week that, despite the concerns about Generative AI (GenAI), the tech community can overcome these problems by putting the human experience at the heart of applications.
“There’s a question about whether or not Generative AI is safe, which really means whether it is operating in a way that is within the parameters you would like,” said Hu. “There’s nothing that inherently stops GenAI from creating content that is offensive, biased, or illegal depending on where you are.
“We should face up to this challenge and work out how to train ChatGPT accurately and generate the right outputs, and develop algorithms to flag things that might be offensive, or might exceed the parameters of what we normally engage with.
“This leads us to a key question around how do you reason with this challenging technology? The way we think about it is Augmented Intelligence because that stops AI being seen as something that’s being done to us, and instead puts the human at the centre of AI and elevates their capabilities.”
Predicting the Future
In the height of a heatwave, tech’s brightest companies and business leaders rubbed shoulders once again with the UK’s political elite for London Tech Week.
Against a backdrop of Generative AI, and a renewed interest in VR and the metaverse, the event played host to start ups, tech leaders, and leaders of parliamentary parties alike.
One of the key themes throughout the event was the boom of use cases businesses are finding for Artificial (or augmented) Intelligence. Speaking in a breakout session Hu went on to demonstrate businesses can utilise AI to drive efficiencies by “working backwards” from the solution to the problem.
“Once you know how to remediate a problem, the natural thing to ask is how can I get ahead of it? With AI, businesses can create the visibility to see problems as they start to occur, or signals that might say something is about to happen, and they can be even more preventative.
“Forecasting has the ability to make a huge difference to businesses. One of the key questions is how do you make your demand forecasting match with your supply? Because this is what unused inventory is, this is what stranded parts are. This is another way businesses can use their overall inputs into the system to be as efficient as possible.
“For any company the specifics are different, but if businesses work their way horizontally along the value chain, and then vertically to see where they have tonnes of data coming in, they have a relatively consistent way of interpreting it, to give some of the insights to generate better quality, better throughput, and better predictability.”
User Error
Despite the advantages, Hu also shared research from Lenovo that found that 43% feel “urgent pressure” to address Artificial Intelligence and Machine Learning as a priority, which is only slightly behind cybersecurity (51%).
Hu also gave the impression that some departments are running before they are walking, citing examples of “engineers [asking ChatGPT] ‘please check this product code is going to be performing well in our runtime” or “someone in marketing drafting a press release for an upcoming launch” which subsequently have been entered into a public ledger.
However Hu ended on a note of positivity, explaining why he’s optimistic about the future of Generative AI.
“If you ask around, there are stories where startups are using Chat GPT and generative models around text like an infinite number of business analysts who can help them make smart decisions.
“And if you talk to data scientists, they are two-or-three times more productive. Talk to software engineers, they are zipping through their engineering tasks like never before.
“By having and thinking about AI as augmented, rather than just artificial, we can emphasise the fundamental humaneness of how this technology can help us elevate.
“That’s why I’m optimistic.”
Keir Starmer – the leader of the UK’s official opposition party – has claimed that politicians need to speed up the deployment of new legislation to regulate artificial intelligence.
Speaking at London Tech Week, the Labour leader outlined differences in his approach to AI including a tougher stance on regulation compared to the “light touch” stance of Rishi Sunak’s government.
Starmer, who was a guest at a fireside chat at LTW, explained that AI brings with it great potential for good, but also warned of the risks involved in adoptions, including spreading misinformation and the threat to jobs.
“We need to put ourselves into a position to take advantage of the benefits but guard against the risks,” he said during a fireside chat at the conference.
Starmer, who hopes to become Prime Minister at the UK’s next General Election (likely to take place in 2024) added: “I think there’s going to have to be an overarching regulatory framework of some sort. The government’s talking about principles in legislation. I think we’re going to have something broader than that.”
Share
