In 2023, AI sent a wake-up call for businesses and consumers, but 2024 marks the turn for politics. The democratization of AI abilities comes at the start of an eventful year that will see more than two billion people, across 50 countries going to the polls. With pundits already raising alarm bells over the technology’s potential to wreak havoc on elections, we need to step back and look at the bigger picture.
Vision 2024, Teneo’s annual CEO & Investor Outlook Survey details that 78% of CEOs are investing in AI, representing a 20-point increase year on year.1 This is against the backdrop of the number one issue facing CEOs this year which is managing upcoming potential political disruption. Combining both of these headwinds could create profound implications for not just business, but democracy at large.
In this piece we’ll delve into the dichotomy of AI’s utility for good and bad actors, highlighting considerations and safeguards.
The Good: Enhancing Engagement and Efficiency
There’s a considerable amount of fear-mongering when it comes to AI’s impact in society and in potential deployment within politics. But solely focusing on the risks (and there are many), will force campaigners to miss out on the benefits of hyper-personalization, campaign efficiencies and a deeper understanding of large data sets to shape policy. Similar to how microtargeting reshaped how campaigns engaged with stakeholders a generation ago; today AI stands ready to reinvent the campaign landscape.
Targeted Communication:
Analyzing large datasets allows campaign planners to identify voter preferences and concerns at scale. This targeted approach can help generate political messages that resonate more effectively with audiences. In addition, using AI tools for linguistic translations can help you reach multi-lingual audiences.
Efficiency and Cost-Effectiveness:
AI-driven automation can reduce the resources required for time-intensive campaign activities. From organizing events to managing social media campaigns and asset creation, AI can streamline operations, allowing for more strategic deployment of resources elsewhere. AI has already been employed for fundraising robocalls in certain U.S. states in the run-up to the presidential election.
Enhanced Digital Media Strategies:
Leveraging AI tools in digital media campaigns can achieve a more impactful online presence. AI tools can optimize content delivery times, personalize messages and monitor public sentiment, taking the guesswork out of online campaign management.
Putting this all together is Ashley, a LLM-powered robocaller enlisted by democratic campaigners in the U.S. to conduct direct voter outreach calls.2
Solving perennial problems for political campaigns – lack of volunteers, high outreach volumes, high rejection rates and the need to personalize the pitch – Ashley speaks 20 languages, is never late to work, can customize the candidate’s pitch and doesn’t take it personally when a voter hangs up. AI that can replace expensive blanket advertisements designed to appeal to a common denominator, tailor messages to individual voters’ concerns and effectively encourage all eligible voters to participate in elections not only stands to help candidates win but also strengthens voters’ engagement with the democratic process.
Sadly, this use case has also been negatively deployed recently by bad actors. During the 2024 New Hampshire Primary, voters received robocalls that mimicked President Biden’s voice and tried to dissuade them from voting.3
Following this, in an example of fast-moving regulatory action, the Federal Communications Commission announced on February 8th that artificial-intelligence-generated voice calls fall under existing “junk calls” law and are prohibited. This move will give states the power to crack down on bad actors, establishing that the use of AI to produce a voice for a robocall constitutes an illegal act.4
The Bad: Ethical and Privacy Concerns
While AI solutions can introduce efficiencies, scale communications and help to deliver a deeper understanding of the voter, there can be ethical concerns with using tools from a generative and data security perspective.
Data Privacy and Security Risks:
The use of AI can potentially lead to the mishandling of personal data. Sensitive information may be misused or fall into the wrong hands, damaging public trust. Those deploying AI must be aware of data obligations when using these tools.
Misinformation and Manipulation:
Generative AI can be exploited to create and spread fake imagery, video and information to manipulate public opinion. This type of misuse poses a threat to the integrity of the democratic process by misrepresenting facts and warping reality. So far we’ve seen political parties use AI to resurrect deceased politicians in Europe to create endorsements from beyond the grave and even generate fake battle imagery to stoke fear.
Potential for Bias:
AI systems are only as unbiased as the data that is used for their training. Inherent biases in data can lead to skewed AI outputs, reinforcing harmful stereotypes and potentially influencing election outcomes.
What does this look like in practice? The human rights advocacy group Freedom House found that generative AI tools had been used in at least 16 countries to distort information on political or social issues. It also found alarming examples coming from political campaigns in the U.S.5
“No matter whether produced for entertainment or to manipulate political beliefs, AI–generated content may shape the information people consume about candidates, voting and the election at large... More broadly, political parties, committees and candidates should refrain from intentionally misrepresenting candidates in advertising that features media generated or manipulated by AI.”6 - Freedom House
The Unknown: Considerations
How do we make sure AI is used in a responsible and ethical way within the political arena in 2024? Here’s some things to consider:
AI Upskilling:
Those using AI tools should understand AI’s capabilities and limitations. This will help them make informed decisions on how and when AI is deployed in organizations. Teneo’s Vision 2024 survey found that 25% of CEOs do not believe they have the right people to power AI adoption in their business. It’s time to invest.
Implementing an AI Code of Ethics:
Businesses should implement AI ethical codes to govern how AI is used – prioritizing truth, accountability and transparency. At Teneo we’ve developed our own code of ethics which guides our consultation and the use of AI for our clients.
Regular Audits and Oversight:
Independent external audits of AI usage can help identify and mitigate risk and biases, ensuring that AI tools are used in a manner that maintains democratic values and complies with ethical codes.
AI Literacy Education:
Regulation and self-governance of AI technologies will only go so far. Some responsibility for discerning fact from AI-generated fiction falls on the shoulders of media outlets and consumers. Surely the same consumers who propelled “Is it Cake?” to Netflix’s top ten list can apply those same interrogation skills to decide “is it fake?”
Conclusion
The rapid adoption of AI during a year of political change offers both extraordinary opportunities and significant challenges for businesses. While AI can revolutionize working efficiencies, it also raises questions about the associated ethical and privacy risks. Navigating this landscape requires a delicate balance, where AI is harnessed responsibly and its pitfalls are mitigated. Following an ethical code and maintaining a commitment to transparency will be vital in order to maintain trust and ensure the integrity of the democratic process.
1https://www.teneo.com/vision2024/
4https://www.msn.com/en-us/news/technology/ar-BB1hZAlX
5https://freedomhouse.org/report/freedom-net/2023/repressive-power-artificial-intelligence
6https://freedomhouse.org/article/democratic-stakes-artificial-intelligence-regulation