The AI Revolution Demands Critical Thinking – Here’s Why

?

Updated for November 2024

?

Originally I wrote this article on AI and critical thinking in March 2023 and felt it was important to update it based on current trends, growth and concerns. If you missed it the first time, here’s the new version.

?

AI was once a distant technology for many, seemingly removed from daily life. But the release of tools like ChatGPT. Gemini, Claude and Bard et al. changed everything. With these tools, ordinary users now interact with powerful generative AI, opening up both exciting possibilities and significant challenges in our complex relationship with information.

?

Today, generative AI models, from ChatGPT and beyond, are transforming how we search, create, and interact with content. By using neural networks, these models can generate unique text, images, and even videos based on massive datasets from books, internet content, and more. While users are often fascinated by these AI capabilities, few truly consider the critical thinking required to engage responsibly with such technology. Educators are also wondering how teaching and learning will evolve when it’s so hard to keep up with the pace of change.

?

Understanding Generative AI and Its Reach

Generative AI models produce new content based on the patterns they find in existing data. For instance, tools like ChatGPT and Midjourney create text or images inspired by user prompts, drawing on vast databases of human knowledge. But it’s critical to remember: these outputs are not replicas; they’re crafted through complex pattern analysis. The capacity of AI to instantly create context-specific content has left many in awe – but as users, we must consider not only what AI can do but also how we respond to and question its outputs. They are also very energy hungry and much has to be done to educate users around their responsibilities in using such technology.

?

From Admiration to Analysis: The Critical Thinking Shift

The initial excitement and frenetic activity around generative AI tools have brought both opportunities and potential pitfalls. For instance, while AI can streamline routine tasks like writing or coding, it also raises the need for users to evaluate content for accuracy and bias. This is not a throwaway comment, it is seminal. Critical thinking, especially in an AI-driven landscape, is not just a skill—it’s essential for discerning fact from fiction.

?

As a former university lecturer, I’ve seen how shortened attention spans and information overload impact students' ability to critically engage with content. Today, users often experience AI-generated content passively, without questioning its sources, reliability, or potential biases. We can find ourselves seduced by the capacity to write content or collate information but are we really learning or interrogating it or just cutting, pasting and sharing? But as educators and individuals, we must cultivate a habit of ?rigorous inquiry when using these tools.

?

Educational Strategies for AI Literacy

Educational institutions are working out how to prepare students for a world of AI-generated content. Many schools and universities incorporate digital literacy and critical thinking exercises focused on evaluating AI-driven information sources. One practical model for building these skills is Bloom’s Taxonomy, which encourages moving from basic comprehension to more advanced stages like analysis and evaluation—both essential when assessing AI outputs. Students learn not just to consume content but to analyse and question its validity, a vital skill in an era where AI is producing a significant portion of online content. It may be a quick way to understand but where does the data come from, who compiled it, who has manipulated that information to be scraped first by AI?

?

Human Manipulation and Bias Introduction: Human-controlled AEO could be used to manipulate AI by flooding it with biased or misleading content. Since AI models, especially those based on machine learning, are trained on vast internet datasets, injecting large volumes of content that mimic authoritative sources or skew facts can push a specific narrative.

?

Updated Concerns for AI’s Sustainability and Ethics

Beyond just critical thinking, AI’s environmental footprint and sustainability have become major concerns. Each query on an AI model involves processing power that requires extensive resources, including rare minerals and high energy costs. Awareness of these hidden impacts should encourage us to use AI responsibly and to consider the broader consequences of its widespread adoption.

??

AI in Gaming and Social Media: A New Frontier for Disinformation

Imagine how these tools might infiltrate digital spaces like gaming, where extremist or propagandist content can be subtly embedded. Tools like Midjourney and Scenario allow for hyper-realistic imagery, and when combined with ChatGPT, they can create violent or manipulative content disguised within games. Normalising conspiracy theories or violent narratives in such platforms could have serious psychological effects on younger audiences.

?

Case Study: The Risks of Disinformation and Misinformation in the 2024 U.S. Election

As generative AI advances, so do the strategies of those who misuse it. In 2021, OpenAI, the Stanford Internet Observatory, and Georgetown University’s Centre for Security and Emerging Technology published a study on AI’s role in disinformation campaigns, warning of the risks AI might bring to democratic processes. With the 2024 U.S. election having just taken place, there is renewed focus on countering the “weapon of mass disruption” scenario, where disinformation campaigns employ sophisticated AI to sway public opinion on a massive scale.

?

This election cycle highlighted ongoing concerns, with platforms having actively prepared for potential uses of AI-powered disinformation to influence voters. Social media saw an increase in AI-generated content designed to mimic legitimate sources, and automated bots amplifying divisive messages. Additionally, advancements in deepfake technology have enabled highly realistic and difficult to distinguish video and audio clips of political figures, posing unique challenges in distinguishing genuine statements from fabrications.

?

The unregulated nature of generative AI tools, such as ChatGPT and deepfake technologies, underscores the risk of polarising content spreading unchecked. The ability to flood social media with AI-generated commentary—whether through hyper-realistic deepfakes or text—raises significant concerns about technological manipulation of voters. Without responsible use, informed regulation, and critical engagement from the public, AI-driven disinformation poses serious challenges to the integrity of democratic processes.

?

Statistical Perspective on AI’s Transformative Scale

The sheer scale of generative AI's impact is illustrated by the increase in the number of users on platforms like ChatGPT, which surpassed 100 million monthly users within its first two months. As of November 2024, ChatGPT has approximately 250 million weekly active users, encompassing both consumer and commercial platforms. This marks a significant increase from the 200 million weekly active users reported in August 2024. ?In the United States, ChatGPT has around 77.2 million monthly active users. ?UK figures are not available.

?

?This rapid growth underscores the urgency of developing critical thinking to navigate the wave of AI content. According to a study from the Pew Research Centre, more than 40% of Americans now report using AI tools regularly. How many have a full understanding of the technology’s limitations or biases?

?

A Call to Action: Strengthening Critical Thinking for AI’s Future

Generative AI is reshaping society and our interaction with knowledge. But as we explore and use these technologies, we must sharpen our critical thinking skills to deal with AI’s transformative effects responsibly. It’s vital to cultivate a mindset that questions what we see, why we see it, and who benefits from its distribution. The traditional ‘gatekeepers’ of information may have disappeared, but the need for individual discernment is greater than ever.

?

Building a Multi-Stakeholder Approach to AI Responsibility

To navigate these complexities, we need a collaborative approach. Governments, educational institutions, tech companies, and individuals must work together to promote responsible AI use and critical thinking education. Some tech companies are already taking steps: for instance, Meta and Google have introduced tools for identifying AI-generated images and video, helping users distinguish between authentic and synthetic media. But these tools are only as effective as the people using them, reinforcing the need for individual responsibility.

?

As we move forward, let’s ensure trust and safety are priorities, not casualties, in the AI revolution. , and it’s up to each of us to engage, understand, and promote a critical approach in the digital age.

?

References

1. Wolleb?k, D., Karlsen, R., Steen-Johnsen, K., & Enjolras, B. (2019). Anger, Fear, and Echo Chambers: The Emotional Basis for Online Behaviour. Social Media + Society, 5(2).
2. OpenAI, Georgetown University’s Centre for Security and Emerging Technology, & Stanford Internet Observatory: Forecasting potential misuses of language models for disinformation campaigns and how to reduce risk.

?