Reframe your m(AI)nd

Our nudge is the Framing Awareness Bot, an AI-based assistant that helps users identify framing in political and societal issues as well as in AI-generated responses. The bot supports users not only in recognizing framing, but also in searching for information, reflecting on complex topics, and making more informed decisions. It analyzes submitted questions or answers for implicit assumptions, evaluative language, and one-sided perspectives. It then highlights which parts represent objective information and which are opinions or interpretations. In addition, the bot encourages users to consider alternative viewpoints, broader context, and reliable sources. To support reflective decision-making, the bot aims to respond as neutrally as possible, provide transparent reasoning, and include sources whenever relevant. Instead of simply reinforcing existing opinions, it helps users critically evaluate information before forming
judgments. The nudge works by not directly restricting behavior, but by subtly encouraging users to engage with information in a more critical and reflective way. This promotes a more conscious and responsible use of artificial intelligence.

Accept the cookie setting to see Panopto content.

Unlock Panopto

What does the topic mean?

Our topic focuses on the interaction between politics, societal issues, artificial intelligence, and framing. More and more people use AI systems such as chatbots to gather information about political and societal topics, form opinions, or prepare decisions. Many users assume that AI provides neutral and objective answers. In reality, however, AI responses are strongly influenced by how prompts are phrased. This phenomenon is called framing. Framing refers to the influence on how people perceive and evaluate a topic based on the way information is formulated or presented. Especially in political and societal contexts, framing can lead to certain perspectives being favored while others are overlooked.
In addition, many AI systems show sycophancy, meaning the tendency to agree with users rather than critically questioning their statements. This can create the illusion that AI responses are neutral, even though they are influenced by language, context, and wording.

Goal of the nudge

The goal of our nudge is to encourage users to interact with AI in a more conscious and reflective way. People should understand that AI is not automatically neutral and that responses can be shaped by language, context, and implicit assumptions. The Framing Awareness Bot aims to strengthen critical thinking, promote perspective diversity, and prevent users from accepting one-sided or biased information without questioning it. In the long term, the nudge aims to improve media literacy and AI literacy in society.

    Needs analysis

    • Growing Use of AI
      More and more people use AI in everyday life, especially for information retrieval, research, and opinion formation.
    • Lack of Awareness of AI Bias
      Many users are unaware that AI responses are influenced by training data, prompts, and linguistic framing.
    • High Relevance in Political and Societal Topics
      Political and societal issues are often emotional, complex, and shaped by conflicting interests. In these contexts, framing can strongly influence perception and opinion formation.
    • Risk of Uncritical Acceptance of AI Responses
      Because AI is often perceived as objective, many people accept its responses without critically evaluating them.

    Cause analysis

    • Framing-Effekt

      People evaluate identical information differently depending on how it is presented or worded.

    • Confirmation Bias

      People tend to seek information that confirms their existing beliefs while ignoring
      contradictory perspectives.

    • Authority Bias

      Statements from perceived authorities are more likely to be accepted. AI is often seen as a neutral and reliable authority.

    • Sycophancy

      Many AI systems tend to reinforce users’ opinions instead of actively introducing counterperspectives.

    • Cognitive Ease

      Simple and easy-to-understand answers are often perceived as more accurate or
      trustworthy, even when they are one-sided.

Target Group

Our main target group includes political institutions, public authorities, as well as educational and consulting organizations that use AI in their daily work. In addition, the nudge is aimed at organizations that use AI in seminars, workshops, or political education programs. Beyond that, the Framing Awareness Bot is relevant for anyone who uses AI to obtain information and wants to learn how to critically evaluate AI-generated responses.

Added value of the nudge

  • Supports Informed Decision-Making

    The bot helps users gather, structure, and reflect on political and societal informat on in
    order to make more informed decisions.

  • Promotes Neutral and Transparent Information Processing

    By distinguishing facts from opinions and providing sources, the bot encourages more
    trustworthy and balanced information evaluation.

  • Strengthens Critical Thinking

    Users become more likely to question AI-generated responses instead of accepting them without reflection.

CONTACT US

Diana Le

diana-duyen.le@stud.hshl.de

Sofia Sianas

sofia.sians@stud.hshl.de

Amelie Reinelt

amelie.reinelt@stud.hshl.de

Lea Beverungen

lea.beverungen@stud.hshl.de