In September 2024, Sam Altman, the CEO of OpenAI, proclaimed, “We have entered the Intelligence Age.” He emphasised the transformative power of deep learning, a subset of artificial intelligence (AI), in learning from massive datasets and its potential to solve the complex problems of our age. His words echo a growing conviction that AI, armed with increasing volumes of data, can help us navigate the intricate and often chaotic challenges we face today.
However, sceptics argue that humans are fundamentally irrational, driven more by emotion and self-interest than reason, and thus limit the utility of AI. While these concerns are valid, they are only part of the story. AI, with its capacity for pattern recognition, forecasting, and providing insightful recommendations, holds the key to a more prosperous world, even amid human irrationality.
Let’s explore how AI can enhance the world, even in the face of human irrationality. We’ll highlight data-driven examples and demonstrate how AI can complement humanity’s emotional and often unpredictable nature.
Irrationality meets intelligence
Human beings are emotional creatures, as political upheavals, misinformation campaigns, and divisive social movements have vividly demonstrated. The recent US election, which saw a significant spread of conspiracy theories and misinformation, underscores that people are often driven by visceral emotions rather than logic or data. Nonetheless, these tendencies do not negate the value AI can offer in creating a more intelligent and effective decision-making framework.
AI is uniquely positioned to cut through emotional biases. As opposed to human beings who bring their past experiences, prejudices, and emotional baggage to each decision, AI systems evaluate problems with a neutral lens. They can process complex datasets with an impartial focus, providing recommendations and predictions untarnished by personal biases. By acting as an objective tool, AI can guide human decision-making in ways that account for but are not overly swayed by our irrational tendencies.
Consider climate change—an issue that is complex, multi-faceted, and undeniably urgent. Despite decades of accumulating evidence, action on climate change has often been delayed due to political wrangling, economic interests, and even outright denial—all reflections of human irrationality. AI can circumvent some of these barriers by providing accurate climate modelling, predictive analytics, and optimisation strategies that help policymakers make informed decisions.
In a 2020 Google report, researchers described how AI models have been used to predict deforestation in the Amazon rainforest. By combining satellite images with machine learning algorithms, AI can identify areas at risk of illegal logging, allowing authorities to intervene before it’s too late. This predictive capability is crucial in a context where political or economic considerations may otherwise delay action. In Germany, AI-enabled systems have also helped to optimise wind turbine efficiency by analysing weather patterns in real-time, resulting in a notable increase in renewable energy production.
Healthcare: AI navigates complexities with ease
The healthcare sector represents another area where human irrationality—such as mistrust in medical systems or biases against new treatments—can lead to poor outcomes. However, AI can help healthcare professionals improve diagnosis, optimise treatment plans, and ultimately enhance patient outcomes, even in the face of human hesitance.
AI models such as IBM’s Watson have demonstrated how AI can assist in diagnosing diseases, including rare cancers, by evaluating a patient’s symptoms against vast medical literature—something no single physician could achieve alone.
In the COVID-19 pandemic, AI played an essential role in tracking virus spread, predicting hotspots, and even assisting pharmaceutical companies in expediting vaccine development. In fact, the vaccine’s rapid development was, in part, thanks to algorithms that helped identify effective molecular compounds in record time.
AI also addresses mental health issues, an area fraught with stigma and misunderstanding. Applications like Woebot and Wysa, which are AI-driven chatbot therapists, provide emotional support to individuals who may feel uncomfortable seeking traditional therapy.
Despite the emotional complexity of mental health, these AI tools have proven effective for many users, providing cognitive behavioural therapy techniques, mood tracking, and supportive dialogue without any judgment. By offering a non-human intermediary, artificial intelligence has successfully overcome barriers like shame and social stigma that often prevent people from seeking help.
Leveraging AI for peace and security
Human irrationality has also led to countless global conflicts, where emotions like fear, anger, and a sense of injustice drive people to violence. Traditional diplomacy has its limits, often subject to political pressures, historical grievances, and the whims of national leaders. AI, on the other hand, can serve as a stabilising influence in international relations by analysing data on socio-economic conditions, public sentiment, and historical conflicts to predict potential flashpoints and recommend interventions.
For instance, the AI for Peace initiative—a collaboration involving the United Nations and various NGOs—has used machine learning models to predict conflicts in African regions based on data related to food scarcity, economic disparity, and historical violence. These insights have allowed for proactive diplomatic interventions and resource allocation, potentially averting conflicts before they spiral out of control.
In Ukraine, AI has been instrumental in predicting Russian troop movements using satellite imagery, allowing the Ukrainian military and its allies to prepare defensive strategies. By providing real-time, reliable data, AI helps mitigate the impact of emotionally charged decisions made under duress. Thus, while AI alone cannot stop conflicts, it provides rational insight that can support and inform human peace-building efforts.
AI has sometimes been criticised for perpetuating inequality, as seen in the controversial case of Australia’s Robodebt programme. This artificial inteligence-driven initiative wrongly accused many welfare recipients of owing debt, causing widespread distress.
It’s essential to acknowledge that AI is not infallible; rather, it reflects the values and biases programmed into it by human developers. However, the key lesson from Robodebt is not that AI is inherently flawed, but that ethical considerations must be integral to its design.
When AI is designed thoughtfully and deployed ethically, it can be a powerful tool to reduce inequities. For instance, India’s Aadhaar programme, which utilises biometrics and AI for identity verification, has helped to streamline welfare distribution, reducing fraud and ensuring that subsidies reach those most in need. The United States has seen similar successes with AI tools for identifying at-risk students, helping schools allocate resources more effectively to support their educational progress.
Reforming the criminal justice system
Human irrationality in the form of prejudice and bias is particularly evident in the criminal justice system, where racial and socioeconomic factors often play a role in sentencing. AI can help mitigate these biases when used correctly. In the United States, risk assessment tools are being used to predict the likelihood of reoffending, and help judges make more informed bail and parole decisions.
A well-known issue with early AI systems in criminal justice was that they learnt from historical data, which already contained systemic biases. This led to unfair predictions that disproportionately affected marginalised communities. Addressing this requires better data collection practices, more diverse development teams, and ongoing audits to ensure fairness. When properly managed, AI can bring a level of consistency and rational evaluation that human judges, often influenced by emotions, may struggle to maintain.
In the UK, for example, the Durham Constabulary has used the Harm Assessment Risk Tool (HART) to predict low-risk offenders and divert them from prosecution, favouring rehabilitation programmes. This approach focuses on reducing reoffending rates, ultimately benefiting both individuals and society. AI’s objective analysis can thus contribute to a more rational, equitable justice system, reducing reliance on subjective human judgment.
Supporting rational public discourse
One of the primary arguments against AI’s efficacy is its role in amplifying misinformation, which can significantly fuel human irrationality. AI-driven algorithms have indeed contributed to the spread of fake news, as seen in the manipulation of social media platforms during elections. However, AI can also be part of the solution in combating misinformation.
AI models developed by companies like Factmata and Logically are being used to identify and flag false information in real-time, helping platforms like Twitter and Facebook reduce the spread of fake news. These tools use natural language processing to analyse news articles and social media posts, identifying misleading content with a high degree of accuracy.
Furthermore, artificial intelligence-driven recommendation systems can be adjusted to prioritise verified information and promote high-quality content. Facebook, for example, has made changes to its news feed algorithm to promote more reliable sources, reducing the visibility of clickbait and misleading headlines.
By providing data-driven insights, AI helps individuals and organisations understand the broader consequences of their actions, offering a more rational basis for ethical deliberation. For instance, companies are increasingly using artificial intelligence to conduct ethical impact assessments before launching new products.
AI can model potential environmental impacts, assess supply chain risks, and even predict social backlash—providing leaders with the information they need to make more conscientious decisions. AI becomes a partner in ethical reasoning, expanding the scope of human considerations without replacing the essential moral compass that individuals and societies must provide.
Human-AI collaboration has already led to remarkable innovations, such as autonomous vehicles that promise to reduce the 1.35 million fatalities caused annually by traffic accidents, the majority of which are due to human error. Here, AI’s rational capabilities compensate for human flaws, helping create safer and more efficient transportation systems.
AI and the future of human flourishing
The fear that AI will lead us to an era dominated by cold rationality devoid of human values—a dystopia imagined by theorists like Theodor Adorno and Max Horkheimer—overlooks the potential for AI to enhance human flourishing. AI is a tool, and its impact depends on how we choose to use it. It can be leveraged for purposes that align with human values: improving healthcare, reducing inequality, mitigating climate change, and fostering peace.
AI is also increasingly being used in creative fields. Tools like OpenAI’s DALL-E and GPT-4 are helping artists, writers, and filmmakers explore new forms of creative expression. These AI systems are not replacing human creativity but expanding its horizons, offering novel ideas and techniques humans can build upon. The interplay between human emotion and AI-generated inspiration exemplifies how rational algorithms and human creativity can coexist and enhance one another.
The Intelligence Age is not a replacement for the age of empathy, emotion, or creativity. Instead, it is a complementary phase that, if guided wisely, can help us address our most pressing challenges. By harnessing AI’s capacity for data-driven reasoning while maintaining our uniquely human values, we can strive towards a future where intelligence and emotion are balanced for the collective good.