This month, Senior Client Executive on our Public Affairs team, Holly Ryan, discusses the rise of Chatbots and AI, and the possible impact on political communications in Ireland.
Politicians are often accused of saying the same thing– “They’re all the same” is a refrain we often hear when discussing our elected representatives. But as chatbots begin to make their mark on political contributions around the world, is this about to become a reality, are politicians really all about to sound the same?
In recent weeks we have seen political figures in the American House of Representatives and in the Australian Parliament using material generated by ChatGPT in parliamentary debates.
These chatbots, which are built using deep learning techniques, learn patterns in language and generate text responses to the tasks submitted.
From political campaigns to policy-making processes, everyday tasks in the political arena could now be carried out in previously unimaginable ways potentially transforming political processes across the globe.
While some hail the technology as a game changer, others warn of the potential ethical and social ramifications of its use.
So, what would Irish politics look like if the use of AI Chatbots takes off?
Speechwriting
As showcased in Australia and the United States, AI technology is already being used in politics to alleviate the often expensive and time-consuming task of speechwriters to produce tailored parliamentary material on a daily, if not hourly, basis.
The technology can be used to analyse a politician's previous speeches and identify the most impactful themes and arguments used to generate material that takes each of these elements into account.
It doesn’t take much to imagine a panicked staffer preparing for Leader’s Questions utilising Chat GPT to gather information on the myriad of issues that may be raised by opposition on any given day.
However, given AI's inability to understand the complex nuances of human language, there are concerns that AI-generated speeches might not always accurately portray the politicians’ personal and emotional opinions accurately.
Measuring Public Opinion
Today, political staff employ AI to analyse engagement with traditional media, and on social media platforms via keyword tracking. While AI can assess large amounts of data across multiple media platforms quickly and automatically, it does not always accurately capture the nuances of human language or sensitive issues.
The problem is this could lead to incorrect or misleading results which do not accurately reflect the views of voters.
Furthermore, at a time where data protection is a non-negotiable element of public processes, using AI to measure public opinion requires the collection and evaluation of personal data, which must be handled with extreme care.
Engagement with Constituents
An area where we may soon see the deployment of AI technology is in responding directly to constituents' queries. This can provide constituents with a timely response while also giving politicians an insight to the recurring themes and issues on voters' agendas. Can chatbots be utilised to help local politicians to better manage their constituency workload? Are we about to see a savvy TD use AI technology to respond to constituency queries and begin the wind down of clinics?
Tempting as this may be, it does of course remove a portion of the direct engagement that politicians have with constituents through email and engagement on social media, replacing it instead with automated responses, that could risk them losing important relationships with reliable voter bases that take years to cultivate.
As a matter of interest, I asked ChatGPT how AI technology could change political processes. This is what it responded:
- Improved decision-making: AI can process vast amounts of data and provide insights to politicians, enabling them to make more informed decisions.
- Personalised political campaigns: AI can analyse individual voter data to create tailored political campaigns and messaging that resonates with specific demographics.
- Political polarisation: AI-powered social media algorithms can reinforce people's existing beliefs, leading to increased political polarisation and potential echo chambers.
- Automated administrative tasks: AI can automate repetitive administrative tasks, such as data entry or report generation, freeing up time for politicians and their staff to focus on other areas.
- Enhanced security: AI can help prevent election interference and hacking attempts by detecting and blocking malicious activities in real time.
When asked to provide an unbiased response to the same question, it provided the same answer...
As the technology continues to advance, it is crucial to consider the ethical implications of its use. But politicians will also have to consider the impact of losing the personal and often nuanced considerations that go into every contribution and how that could impact them at the polls.