Fri. Feb 23rd, 2024
    Robocall Manipulating Elections with Deepfake Technology

    A recent robocall attempting to discourage New Hampshire residents from voting in the Democratic presidential primary has experts buzzing about the use of artificial intelligence (AI) to spread disinformation. Disguised as President Joe Biden, the robocall, which has been deemed an “unlawful attempt” to suppress voters, is suspected to be a deepfake – a fake audio or video created using AI technology.

    While the origin of the call remains unknown, disinformation experts and AI researchers agree that it bears the hallmarks of a deepfake. Ben Colman, CEO of Reality Defender, a company specializing in media file analysis, stated that all indications point towards the call being AI-manipulated. It is highly likely that the voice on the call is not actually President Biden’s, but an imitation created through widely accessible AI programs.

    AI algorithms capable of mimicking someone’s voice convincingly are readily available as phone apps or online services, and they typically only require a small amount of sample audio to replicate the individual’s speech patterns. This makes it exceedingly easy for malicious actors to use AI-generated deepfakes to mimic politicians and manipulate public opinion.

    While there are often telltale signs of deepfakes, such as unnatural cadence or robotic inflections, developers are continually refining the technology, making it increasingly difficult to detect fake content. Lindsay Gorman, an expert in emerging technologies and disinformation, notes that visual cues, such as unnatural eye movements, were once reliable markers of deepfakes but have since been addressed by improved algorithms.

    The incident has prompted concerns from lawmakers like Senator Richard Blumenthal, who introduced AI legislation in the Senate. Blumenthal believes the deepfake incident should serve as a wake-up call, highlighting the potential dangers of disinformation spread through AI technology.

    While federal law prohibits attempts to hinder voter participation, deceptive use of AI remains largely unregulated. Although calls for the Federal Election Committee to regulate deepfake ads have been made, no concrete steps have been taken thus far.

    Mekela Panditharatne, senior counsel at the Democracy Program at the Brennan Center for Justice, points out that while deepfake technology may be new in the context of U.S. presidential primaries, robocalls have long been employed by deceptive actors to spread false information and suppress votes. The introduction of voice-generation AI amplifies the threat by providing fraudsters with a more sophisticated tool.

    As the battle against disinformation evolves, it is crucial for lawmakers, technology experts, and the public to remain vigilant and proactive in addressing the challenges posed by AI-manipulated content.

    Frequently Asked Questions about AI-Generated Deepfakes and Disinformation

    Q: What is a deepfake?
    A: A deepfake refers to a fake audio or video created using AI technology that convincingly mimics someone’s voice or appearance.

    Q: What is the recent robocall incident in New Hampshire?
    A: The recent robocall in New Hampshire was an unlawful attempt to discourage residents from voting in the Democratic presidential primary. The call, disguised as President Joe Biden, is suspected to be a deepfake.

    Q: How can AI algorithms create convincing voice imitations?
    A: AI algorithms capable of mimicking someone’s voice convincingly are readily available as phone apps or online services. They typically require only a small amount of sample audio to replicate the individual’s speech patterns.

    Q: What are the concerns about deepfakes?
    A: Deepfakes can be used to manipulate public opinion, spread disinformation, and deceive people. With advancements in technology, it is becoming increasingly difficult to detect fake content.

    Q: What are some signs of deepfakes?
    A: While there may be telltale signs of deepfakes, such as unnatural cadence or robotic inflections, developers are continually improving the technology to make them more convincing. Visual cues, such as unnatural eye movements, were once reliable markers but have since been addressed by improved algorithms.

    Q: How are lawmakers responding to the threat posed by deepfakes?
    A: Lawmakers, like Senator Richard Blumenthal, have expressed concern about the potential dangers of disinformation spread through AI technology. While federal law prohibits attempts to hinder voter participation, deceptive use of AI is largely unregulated. Calls for the Federal Election Committee to regulate deepfake ads have been made, but no concrete steps have been taken so far.

    Q: How does the use of AI amplify the threat of robocalls spreading disinformation?
    A: While robocalls have long been used to spread false information and suppress votes, the use of AI-generated voice imitations makes them more sophisticated tools for deceptive actors to utilize.

    Definitions:
    – Deepfake: A fake audio or video created using AI technology that convincingly mimics someone’s voice or appearance.
    – Disinformation: False or misleading information spread intentionally to deceive people.
    – AI: Artificial Intelligence, the simulation of human intelligence in machines that are programmed to think and learn like humans.

    Related Links:
    Reality Defender: The website of Reality Defender, a company specializing in media file analysis.
    Brennan Center for Justice: The Brennan Center for Justice, a think tank focused on democracy and justice issues.