> Markdown version of [/videos/1192-the-ai-elections-how-technology-could-shape-public-sentiment](https://www.wearedevelopers.com/videos/1192-the-ai-elections-how-technology-could-shape-public-sentiment). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# The AI Elections: How Technology Could Shape Public Sentiment

Are deepfakes the ultimate threat to global elections? The real danger isn't advanced AI, but our shrinking attention spans that let synthetic manipulation slip by unnoticed.

- **Speakers:** [Martin Förtsch](https://www.wearedevelopers.com/@martin-fortsch), [Thomas Endres](https://www.wearedevelopers.com/@thomas-endres)
- **Event:** World Congress 2024
- **Published:** August 29, 2024
- **Duration:** 30:02
- **URL:** https://www.wearedevelopers.com/videos/1192-the-ai-elections-how-technology-could-shape-public-sentiment

## Summary

As roughly half the global population heads to the polls, the rapid advancement of generative AI introduces unprecedented risks to public sentiment and democratic processes. Technologies capable of generating deepfakes, synthetic audio, and automated social media narratives can easily be weaponized to manipulate voters. The most dangerous phase for any such technology is when it is already in use by malicious actors but remains largely unknown to the general public.

Creating convincing real-time video deepfakes involves a multi-step pipeline: utilizing tools like MediaPipe for facial landmark mapping, applying face segmentation, and using auto-encoders to compress and decode facial features via a latent space. Beyond complex video manipulation, malicious actors leverage language models like GPT-2 and audio generators to deploy automated social media bots and synthetic voice clones. Even more concerning are "shallow fakes"—simple techniques like slowing down video audio or presenting genuine images out of context—which require no advanced AI but are highly effective at exploiting human cognitive biases and the preference for novelty.

Unfortunately, automated fake detection algorithms frequently fail when confronted with undocumented or proprietary generation methods. While a live video deepfake illusion can be temporarily broken by asking a participant to pass a hand over their face, combating the broader spread of misinformation demands robust media literacy, critical thinking, and rigorous source-checking. Ultimately, the greatest vulnerability is not the technology itself, but the human attention span; when users only pause for a few seconds to consume content, the friction necessary to identify deception is entirely lost.

**Keywords:** AI elections, generative AI threats, real-time deepfakes, social media bots, public sentiment manipulation, shallow fakes, synthetic audio generation, automated misinformation, facial segmentation, auto-encoder latent space, deepfake detection algorithms, cognitive biases in media, media literacy countermeasures, GPT-2 fine-tuning, video manipulation strategies

## Chapters

1. **Impact of generative AI on democratic elections** (00:19) — How AI-generated images and media manipulation scale up to influence global democratic structures.
1. **Technical pipeline for real-time video deepfakes** (08:51) — The process of face detection, segmentation, and autoencoder generation to swap facial structures in real time.
1. **Exploiting deepfakes in live video conferencing** (14:17) — Using real-time voice and video manipulation to deceive targets in live meetings and extract funds.
1. **Analyzing visual artifacts in synthetic media** (16:34) — How to spot current generation deepfakes by observing unnatural skin patterns, glitches, and lagging facial features.
1. **Automating social media bots with large language models** (17:45) — Training localized text generation models on platform scraping to flood discussions with autonomous comments.
1. **Generating and evaluating synthetic voice cloning** (21:33) — The minimal training data requirements for models capturing speech cadence and intonation to produce synthetic audio clips.
1. **Evaluating automated detection tools and shallow fakes** (22:55) — Why algorithmic detection struggles with novel fakes and how simple pitch alterations create convincing false narratives.
1. **Mitigating misinformation and navigating human attention** (26:46) — Guidelines for manually verifying digital content sources amidst the challenges of rapid social media consumption.
1. **Using social network topology for bot detection** (28:25) — The challenge of identifying automated accounts based on standalone text versus network interaction histories.

## Related Moments

- [Generative AI accelerating the production of online deception](https://www.wearedevelopers.com/videos/1449-the-four-horsemen-of-the-information-apocalypse-taming-the-wild-west-of-information-overload) (from "The Four Horsemen of the Information Apocalypse: Taming the Wild West of Information Overload")
- [Audience questions on deepfake detection, regulations, and ethics](https://www.wearedevelopers.com/videos/624-the-shadows-that-follow-the-ai-generative-models) (from "The shadows that follow the AI generative models")
- [Introducing deepfakes and the controversy of manipulated media](https://www.wearedevelopers.com/videos/180-deepfakes-in-realtime-how-neural-networks-are-changing-our-world) (from "Deepfakes in Realtime - How Neural Networks Are Changing Our World")
- [Addressing the authenticity of AI translated video content](https://www.wearedevelopers.com/videos/1819-wearedevelopers-live-yes-css-can-do-that) (from "WeAreDevelopers LIVE - Yes, CSS Can Do That!")
- [Exploring computer-generated content and modern deepfake capabilities](https://www.wearedevelopers.com/videos/1187-deep-fakes-the-lies-we-can-t-see) (from "Deep Fakes: The Lies We Can’t See")
- [Regulating deepfake content and AI transparency in Europe](https://www.wearedevelopers.com/videos/2138-wearedevelopers-live-node-and-package-security) (from "WeAreDevelopers LIVE - Node and Package Security")

## Related Articles

- [The Rise of Disinformation on the Web](https://www.wearedevelopers.com/magazine/539-the-rise-of-disinformation-on-the-web)
- [How machine learning can help us tell fact from fiction](https://www.wearedevelopers.com/magazine/509-how-machine-learning-can-help-us-tell-fact-from-fiction)
- [The importance of technology in modern democracy](https://www.wearedevelopers.com/magazine/496-the-importance-of-technology-in-modern-democracy)
- [WWC24 Talk - Scott Hanselman - AI: Superhero or Supervillain?](https://www.wearedevelopers.com/magazine/469-wwc24-talk-scott-hanselman-ai-superhero-or-supervillain)

## Related Jobs

- [AI Software Engineer (Germany)](https://www.wearedevelopers.com/jobs/48317-ai-software-engineer-germany) at **Sunhat**
- [Staff Developer Advocate, GitHub Security Lab](https://www.wearedevelopers.com/jobs/ext/1921051-staff-developer-advocate-github-security-lab) at **GitHub**
- [AI & Machine Learning Engineer (all genders)](https://www.wearedevelopers.com/jobs/48217-ai-machine-learning-engineer-all-genders) at **msg**
- [Senior AI Agent Software Engineer (Go, Python) (m/f/x)](https://www.wearedevelopers.com/jobs/48277-senior-ai-agent-software-engineer-go-python-m-f-x) at **Dynatrace**
- [Security Architect - AI](https://www.wearedevelopers.com/jobs/ext/1581899-security-architect-ai) at **ZEISS Group**
- [AI Operations Manager (all genders)](https://www.wearedevelopers.com/jobs/48263-ai-operations-manager-all-genders) at **envelio**