AI holds lots of promise, but also a dark side
Emerging artificial intelligence technology such as ChatGPT could hold many benefits for society, but could also pose a threat to personal privacy, according to panellists at the International Tech Summit 2023.
Caroline O’Brien, chief data officer and head of product at Afiniti, said: “AI is usually created using data. There are ways in which the data that is generated can be harmful, or can impact someone’s rights to privacy.”
Speaking during a panel called AI Unleashed: Amplifying Human Potential, Dr O’Brien said that although language models such as ChatGPT scrape “massive amounts” of data from publicly available sources, such as books or Reddit forums, there is still the risk that it could provide personal data when prompted. Much of the data it finds on the internet is uncurated.
She was also concerned about employees in customer-facing enterprises cutting and pasting consumer data into the system to experiment.
“At the moment ChatGPT is using the information you put into it for product development, rather than learning, but that may change in the future,” Dr O’Brien said. “According to their website, they may use it for other purposes such as selling to third parties.”
ChatGPT is a powerful artificial intelligence chatbot released last November. People can ask the chatbot questions and it will come up with a response scraped from public sources in books and across the internet. ChatGPT 3 has about 170 billion parameters, and ChatGPT 4, the most recent version, has even more.
There is also concern about misinformation.
“We put my name into ChatGPT and it gave out scurrilous information,” Dr O’Brien said.
ChatGPT also mistakenly labelled panellist Dmitry Mnushkin, as a Russian oligarch instead of chief technology officer and cofounder of Swiftly Holdings.
Last week, Italy became the first Western country to block ChatGPT over privacy concerns and Canada has opened an investigation into the chatbot.
But speakers at the International Tech Summit were largely excited about AI’s trajectory.
“I see it as amplifying our potential and accelerating what we can do,” said moderator Michael Branco, founder of Celeste Ventures. “I read a Forbes article this weekend that said AI could add 7 per cent to the gross domestic product of the United States.”
Carika Weldon, chief executive officer and founder of CariGenetics, said AI will have a positive impact on genetic testing in the science arena.
“We might see in our hospital in the next few years where clinicians are able to use a tool that uses AI machine learning to analyse genetic information on the spot,” Dr Weldon said. “When it comes to your health, time is everything.”
Mr Mnushkin said in an AI-driven call centre insurance companies could handle claims during major events much faster.
“Now there are all kinds of possibilities that we are excited about,” Mr Mnushkin said.
The conversation around AI continued at the summit into a later panel on digital privacy and cybersecurity, moderated by Polina Branco, executive board member at Celeste Ventures.
Liat Shetret, director of global policy and regulation at Elliptic, said that as a compliance professional she was worried.
“Do we really want more unicorns that are breaking boundaries and flying high?” she asked. “I want to go back to camels. I want slow and steady. I want controlled growth that has hard wiring for compliance concerns, risk managing as we can. Some of that comes back to doing due diligence and considering risks and understanding the infrastructure that needs to go into cybersecurity management concerns, and all these other pieces.”
Samantha Simms, CEO of The Information Collective, said with AI we are moving into uncharted territory.
“Right now we, for the first time, have been given full public access to something that most countries are trying to regulate but cannot get their arms around,” she said.
She questioned why chatbots such as ChatGPT have been released to the public, just as the European Union and the United Kingdom were looking at regulatory frameworks for AI.
Marisa Stones, senior analyst at the Government’s Privacy and Pati Unit, said since the pandemic began there has been an “explosion of concern” over how personal information is being used.
“Following the principles of privacy by design and security by design is very important in the development of new pieces of software and new applications of technology,” Dr Stones said.
“In addition to the privacy and security design, I like to lean on cultural values by design,” ChaVon Clarke-Joell, the assistant privacy commissioner, said. “That is the foundation that will keep your business and yourself aware of change.”
The fourth annual International Tech Summit 2023 continues at the Hamilton Princess & Beach Club today and tomorrow.
2. Please respect the use of this community forum and its users.
3. Any poster that insults, threatens or verbally abuses another member, uses defamatory language, or deliberately disrupts discussions will be banned.
4. Users who violate the Terms of Service or any commenting rules will be banned.
5. Please stay on topic. "Trolling" to incite emotional responses and disrupt conversations will be deleted.
6. To understand further what is and isn't allowed and the actions we may take, please read our Terms of Service