Which is the Best Artificial Intelligence? A Test on Nine of Them

How will AI change our views of the future?

Every Artificial Intelligence is an instrument to look into the world. Every single one has different capabilities, filters, and biases. In the future, we’ll see the world mainly through AI “eyes” — what kind of world are we going to see? In this post, I tested several AIs to see how they approached the same problem with included a “trap” to see how in depth they could understand the question. Most of them failed to detect the trap, but some showed remarkable insight. ChatGPT was by far the best, but Deepseek and Grok did well, too.

Not so long ago, as an ordinary person, what you knew about the world was mainly what you could see and experience in your village. About the rest of the world, you relied mostly on mythical stories of heroes and monsters. If you were living in Europe, on Sundays you also heard about events said to have happened in Palestine many centuries before.

Today, almost everything you know about the world is “filtered” by entities on which you have little or no control. Apart from the very limited section that you can see with your eyes and touch with your hands, you rely on that humongous blob that we sometimes call the “media.”

Over a couple of centuries or so, the interface has evolved at a breathtaking pace. Newspapers appeared in the 19th century, and governments made an enormous effort to diffuse literacy among the population. It gave governments the power to tell people what they should know and think. In the 20th century, there came the radio and then the TV. The radio allowed people to hear many more voices than the old press. During WW2, Germans listened to the Londoner Rundfunk Radio, Italians to Radio Londra, the British to Lord Haw-Haw, and the American troops in Asia listened to Tokyo Rose. It was forbidden, but plenty of people did it. Then, TV took over under strict government control.

Then, there came the Internet. Initially, the concept of “Internet Neutrality” raised hopes of unbiased information to everybody, but that was a brief moment. Neutrality was soon abandoned, leaving providers able to slow down or throttle traffic to competitors’ websites, while the dominant search engines could easily obscure or demote information they judged not useful. Today, the Web is controlled by obscure forces that decide what you are supposed to see or not to see, not unlike the old Soviet press and TV.

And then Artificial Intelligence came. That’s something so new that we still can’t be sure of what its impact will be. What we know, though, is that AIs are rapidly becoming our interface with the real world. Are they to be trusted? AIs, in principle, are under the control of their creators, but they depend on very large databases; they do not use search engines, and how do creators control what their creatures know and what they don’t know? The powers that be (the PTB) are just humans with all the limits of human beings.

We still don’t know how our relationship with AIs will evolve and change. Surely, they provide tremendously powerful tools to bypass propaganda and government control. The question is whether we will be given the possibility to use these tools and how we will be able to manage them. We could be evolving toward a “radio” broadcasting system in which you have a spectrum of AIs available which you could use to evaluate what you are told. Or, we could evolve toward a “TV” broadcasting system in which you will have just a small number of AIs to choose from, all tightly controlled by the government and telling you all the same things.

As usual, the future will be what it will be. In the following, I am reporting the data I obtained from a test I made with a question posed to several AIs.

The question

I asked several general AIs the following question:

Did the Club of Rome make wrong predictions with their 1972 study “The Limits to Growth”?

It is a question on a subject I know well; I wrote an entire book on it, “The Limits to Growth Revisited” (2011). So, I could frame the question as a trap: the Club of Rome was accused of having made wrong predictions, but the “Limits to Growth” report didn’t contain predictions. It was a typical smear campaign carried out for political purposes starting in the late 1970s. So, the question was whether AIs could detect the trap or would fall into it.

The result was that most of the current AIs fell into the trap and provided an answer in line with the conventional opinion, still based on the old smear campaign. Some gave rather shallow answers, e.g., Copilot and Claude, and some (e.g., Grok) provided more nuanced answers. Deepseek did very well, although it provided a rather synthetic answer. ChatGPT was by far the best. Right on target, bypassing decades of propaganda.

Among other things, Claude made some glaring mistakes, such as writing that “ they predicted several key minerals would be depleted by the 1990s-2000s, which didn't happen.” which is simply false. Another mistake was to say that “increasing CO2 emissions and rising global temperatures align fairly well with what has occurred.” Which is something that just doesn’t exist in the 1972 report: at that time, the correlation of CO2 and global warming was not yet clear.

Note also that the subject is complex, and it would need much more extensive testing. For instance, in a previous test on a different kind of question — “calculate the expected temperature of the Mesozoic”— Deepseek did much worse than Claude. So, it is a continuously evolving situation. Different question may completely change the answer to the question of which AI is best. But we need to keep track of what goes on.

The answers

Here are the reports for nine general-purpose AIs: Copilot, Claude, Perplexity, Mistral, Kimi, Gemini, Grok, Deepseek, and ChatGpt. The answers are reported here in an approximate order that goes from worst to best.

______________________________________________________________________

Copilot

The Club of Rome's 1972 study, "The Limits to Growth," used computer modeling to project the consequences of exponential economic and population growth with finite resource supplies. The study examined five key factors: population growth, industrial output, pollution, food production, and resource depletion.

Critics argue that some of the study's predictions have not come to pass, particularly the dire scenarios of global collapse by the mid-21st century. However, supporters point out that the study's broader message about the unsustainability of continuous growth and the need for systemic changes remains relevant.

In essence, while some specific predictions may have missed the mark, the study's core warnings about environmental and resource limits continue to resonate with many experts today.

__________________________________________________

Claude:

The "Limits to Growth" predictions have been extensively debated. Looking at their key predictions:

The report warned about resource depletion, pollution, and population growth leading to societal collapse potentially by mid-21st century if growth patterns continued unchanged. Some specific predictions:

1. Their resource depletion forecasts were too pessimistic. They significantly underestimated technological advances in resource extraction and efficiency. For example, they predicted several key minerals would be depleted by the 1990s-2000s, which didn't happen.

2. However, their pollution and environmental impact predictions have proven more accurate. Their models showing increasing CO2 emissions and rising global temperatures align fairly well with what has occurred.

3. Their population growth estimates were high - they didn't fully account for how development leads to declining birth rates.

4. Their overall system dynamics model showing interactions between population, industrialization, food production, and environmental degradation has held up better than individual specific predictions.

The report's core insight - that unlimited exponential growth isn't sustainable on a finite planet - remains relevant. But they underestimated humanity's ability to innovate and adapt in the short-medium term, while their longer-term concerns about environmental limits and system collapse remain debated.

Interestingly, a 2014 study by Graham Turner at the University of Melbourne compared their "business as usual" scenario against 40 years of actual data and found the broad trends aligned fairly well with reality, though specific numbers were often off.

________________________________________________________________

Perplexity

Answer

The Club of Rome's 1972 report, The Limits to Growth, has been a focal point of debate regarding its predictions about resource depletion and environmental collapse. The report utilized computer simulations to forecast that if current trends in population growth, industrialization, resource use, and pollution continued, the world would face severe consequences within a century. This message resonated widely, especially during the oil crisis of 1973, leading to its status as a best-selling environmental book15.

Predictions and Criticisms

Critics have pointed out that many specific predictions made in The Limits to Growth did not materialize as anticipated. For example, the report suggested that the world would run out of various resources, including oil and several metals, by specific dates in the 1980s and 1990s. These predictions were later deemed overly pessimistic as technological advancements and resource management strategies led to increased efficiency and new discoveries23. Notably, some critiques emphasize that the authors underestimated human ingenuity and adaptability in addressing resource challenges2.Despite these criticisms, supporters argue that the fundamental thesis of the report—that indefinite economic growth on a finite planet is unsustainable—remains valid. Studies conducted decades later have shown that many trends predicted by the report align with ongoing environmental issues and resource management challenges today. For instance, research by Gaya Herrington in 2020 indicated that data since the 1970s supports the scenarios outlined in The Limits to Growth, suggesting potential societal collapse if current practices continue unchecked16.

Evolving Perspectives

In response to criticisms, members of the Club of Rome have clarified that their intention was not merely to forecast doom but to provoke a reevaluation of growth paradigms. Aurelio Peccei, one of the founders, noted that the report aimed to challenge complacency regarding growth trends and encourage proactive solutions to global inequities and environmental degradation45. Over time, the Club has acknowledged the need for sustainable growth rather than outright cessation of growth.

Conclusion

In summary, while specific predictions from The Limits to Growth may not have been accurate or timely, the broader implications regarding sustainability and resource limits have gained renewed relevance. The discourse surrounding this report continues to influence contemporary discussions on economic growth and environmental stewardship, highlighting a complex legacy that intertwines alarmism with critical foresight about humanity's relationship with its planet.

________________________________________________________________

Mistral

he "Limits to Growth" report, published in 1972 by the Club of Rome, has been a subject of debate and scrutiny since its release. The report used computer simulations to model the consequences of exponential growth in the world population, industrialization, pollution, food production, and resource depletion. Here are some key points regarding the predictions and their accuracy:

1. Methodology and Scenarios: The report presented several scenarios based on different assumptions about resource availability, technological progress, and environmental policies. The most pessimistic scenarios predicted a collapse of global systems within a century due to resource depletion and environmental degradation.

2. Criticisms: Critics have argued that the report's predictions were overly pessimistic and did not account for technological advancements, market adaptations, and policy changes that could mitigate the impacts of growth. Some have pointed out that certain resources, like oil and minerals, have not been depleted as quickly as predicted, and that new technologies have emerged to address some of the challenges.

3. Accuracy of Predictions: While some specific predictions about resource depletion have not materialized as quickly as the report suggested, other aspects of the report have been more accurate. For example, the report highlighted the potential for environmental degradation and climate change, which are now recognized as significant global issues.

4. Long-Term Perspective: The report was intended to provoke thought and discussion about the long-term sustainability of current growth patterns rather than to make precise predictions. It aimed to raise awareness about the potential consequences of unchecked growth and the need for sustainable development.

5. Updates and Reassessments: The original report has been followed by several updates and reassessments, including "Beyond the Limits" (1992) and "Limits to Growth: The 30-Year Update" (2004). These updates have refined the original models and incorporated new data and insights.

In summary, while the "Limits to Growth" report may not have been entirely accurate in its specific predictions, it has been influential in raising awareness about the potential long-term consequences of unchecked growth and the need for sustainable development. The debate continues about the accuracy of its predictions and the relevance of its warnings in the context of current global challenges.

Kimi

The 1972 "Limits to Growth" report by the Club of Rome has been a subject of much debate over the years, with some of its predictions proving to be more accurate than others. Here are some key points:

Accurate Predictions

• Environmental Pressures: The report warned that continued exponential growth in population, industrialization, pollution, and resource depletion would lead to severe environmental and social pressures. This has largely come to pass, with increasing concerns about climate change, resource scarcity, and ecosystem degradation.

• Need for Sustainable Practices: It emphasized the need for humanity to shift towards more sustainable practices to avoid collapse. This has been increasingly recognized in recent decades, with growing efforts towards renewable energy, resource conservation, and sustainable development.

Less Accurate Predictions

• Timing of Collapse: The report suggested that without significant changes, a collapse of industrial production and population decline could occur by the mid-21st century. While there have been significant challenges, a complete collapse has not yet occurred, and some argue that technological advancements and policy changes have mitigated some of these risks.

• Specific Resource Depletion: The report predicted that certain non-renewable resources would be depleted much faster than they have been. For example, it underestimated the impact of technological advancements in resource extraction and the discovery of new reserves.

Overall Impact and Legacy

Despite some inaccuracies, the report has had a profound impact on environmental and sustainability discourse. It helped to bring global attention to the interconnectedness of economic, social, and environmental systems and the need for sustainable development. The Club of Rome continues to advocate for policies and practices that align with the principles of sustainability and resilience.

In summary, while not all predictions in the "Limits to Growth" report have come to pass exactly as envisioned, its core message about the need for sustainable practices and the potential consequences of unchecked growth remains highly relevant today.

______________________________________________________________________

Gemini

The 1972 "Limits to Growth" report by the Club of Rome used computer modeling to explore the potential consequences of unchecked economic and population growth on a planet with finite resources. While the report has been influential, its predictions have been subject to criticism and debate.

Arguments that the report's predictions were wrong or flawed:

• Technological progress: The report did not fully account for the potential of technological advancements to increase resource efficiency, develop substitutes, and mitigate pollution.

• Resource scarcity: Some argue that resource scarcity has not materialized to the extent predicted by the report, due to factors like improved extraction techniques and recycling.

• Oversimplification: Critics suggest that the model oversimplified complex global systems and did not adequately capture regional variations and socioeconomic factors.

• Pessimistic bias: Some view the report as overly pessimistic, focusing on potential negative outcomes while underestimating human adaptability and innovation.

Arguments that the report's predictions were accurate or still relevant:

• Resource depletion: Concerns about resource depletion remain valid, particularly with increasing consumption in developing countries and the environmental impact of extraction.

• Environmental degradation: The report's warnings about pollution and environmental degradation have become increasingly urgent with issues like climate change and biodiversity loss.

• Unsustainable growth: Critics acknowledge that the report highlighted the potential unsustainability of continuous economic growth based on finite resources.

• Call for action: Even if the specific predictions were not entirely accurate, the report served as a crucial wake-up call about the interconnectedness of human activities and planetary limits.

Overall:

The "Limits to Growth" report was a pioneering effort to model global systems and explore potential future scenarios. While some of its specific predictions have not fully materialized, it raised important questions about resource constraints, environmental impacts, and the nature of economic growth. The report remains a valuable contribution to discussions about sustainability and the need for responsible stewardship of the planet.

It is important to note that the Club of Rome has published several follow-up reports and updates to "Limits to Growth," acknowledging some of the critiques and incorporating new data and insights. These later reports have often emphasized the need for policy changes and international cooperation to address global challenges.

In conclusion, whether the "Limits to Growth" report was "wrong" depends on one's interpretation of its predictions and the extent to which one prioritizes specific outcomes versus the broader message about sustainability. While the report may not have perfectly predicted the future, it has undoubtedly stimulated important conversations and actions regarding the relationship between humanity and the planet.

______________________________________________________________

Grok

The "Limits to Growth" report by the Club of Rome, published in 1972, has been a subject of considerable debate over the years. Here are some key points to consider regarding the accuracy and reception of its predictions:

Predictions and Methodology:

• Scenario-based Predictions: The report used a computer model named World3 to simulate different future scenarios based on current trends in population, industrial output, pollution, food production, and resource depletion. It did not make singular predictions but rather explored various possible outcomes if certain trends persisted without change.

• Main Scenario: The scenario that attracted the most attention suggested that if trends continued unchanged, economic growth would peak and then decline by the mid to late 21st century, leading to a collapse in population and industrial capacity due to resource depletion and environmental degradation.

Criticisms:

• Overestimation of Resource Scarcity: Critics argue that the report overestimated how quickly resources would be depleted, not accounting for technological innovations, new discoveries, and improvements in efficiency that would extend resource availability.

• Underestimation of Human Adaptability: The model did not adequately consider human ingenuity and policy changes that could mitigate negative trends or adapt to new conditions.

• Economic and Political Models: Some critics pointed out that the economic and political systems were modeled too simplistically, not fully capturing the complexity of global responses to crises.

Defenses and Reassessments:

• Long-term Trends: More recent analyses, like the 2004 and 2014 updates by Graham Turner, suggest that the data from the 1970s has tracked surprisingly closely with historical data through to the early 21st century, particularly in terms of population growth and resource use.

• Environmental and Climate Considerations: The report was somewhat ahead of its time in linking economic growth with environmental degradation, a connection that has become more evident with climate change discussions.

• Interpretation vs. Prediction: The Club of Rome intended the report not as a definitive prediction but as a warning or a call to action for systemic changes. The scenarios were meant to be illustrative of potential outcomes if no changes were made.

Conclusion:

• Not Entirely Wrong, Nor Entirely Right: It would be incorrect to say the report was "wrong" in all its predictions. Instead, it highlighted potential paths based on unaltered trends, which in some areas have been mitigated or altered by human actions, technology, and policy.

• Ongoing Relevance: The core message about the limits of exponential growth on a finite planet remains relevant, especially in the context of sustainability, climate change, and resource management.

In summary, while specific predictions from "Limits to Growth" might not have come to pass as precisely forecasted, the report served as an influential catalyst for discussions on sustainability, environmental policy, and economic models, which continue to be relevant today.

_________________________________________________

Deepseek

The 1972 study "The Limits to Growth," commissioned by the Club of Rome and conducted by a team of researchers at MIT, used computer modeling to explore the long-term consequences of exponential economic and population growth in a world with finite resources. The study highlighted the potential for global overshoot and collapse if trends in industrialization, population growth, pollution, food production, and resource depletion continued unchanged.

Key Points of the Study:

• The model projected that if current trends persisted, the global system would likely experience a decline in resource availability, industrial output, and food production, leading to a collapse in population and living standards sometime in the 21st century.

• It emphasized the interconnectedness of global systems and the risks of exceeding Earth's carrying capacity.

Criticisms and Controversies:

1. Misinterpretation of Predictions: Critics often accused the study of making specific, doomsday predictions, but the authors explicitly stated that their goal was to explore scenarios, not to predict exact outcomes. They highlighted the potential for policy changes to alter trajectories.

2. Technological Optimism: Some critics argued that the study underestimated human ingenuity and technological advancements, such as improvements in resource efficiency, renewable energy, and agricultural productivity.

3. Data and Model Limitations: The model relied on assumptions and data available in the early 1970s, which some argued were incomplete or outdated. Critics questioned the accuracy of the projections over such a long timescale.

Subsequent Evaluations:

• In the decades since its publication, several studies have revisited the "Limits to Growth" scenarios. Some have found that the world is broadly following the "business-as-usual" trajectory outlined in the study, particularly in terms of resource depletion, pollution, and environmental degradation.

• For example, a 2014 study by Graham Turner compared the original model's projections with real-world data and found that trends in population, industrialization, and resource use were consistent with the study's "standard run" scenario, which predicted a collapse in the mid-21st century.

Conclusion:

The Club of Rome's "Limits to Growth" study was not a definitive prediction but a warning about the potential consequences of unchecked growth. While some of its specific projections have been debated, its core message—that infinite growth on a finite planet is unsustainable—remains relevant. The study has been praised for raising awareness about global environmental challenges and inspiring further research into sustainable development. Whether its predictions were "wrong" depends on how one interprets its purpose and the extent to which humanity heeds its warnings.

____________________________________________________

ChatGpt:

The Limits to Growth report, published by the Club of Rome in 1972, was not so much about making precise predictions as it was about modeling potential future scenarios based on different assumptions. The report used the World3 computer model to simulate how continued economic and population growth might interact with finite planetary resources.

Were the Predictions Wrong?

• Not Exact Predictions: The authors explicitly stated that their work was not a set of predictions but a set of scenarios that depended on policy choices.

• General Trends Have Held Up: Some studies have compared real-world data with the scenarios modeled in Limits to Growth, and many suggest that trends in resource use, pollution, and population growth have followed a trajectory close to what the report called the "business-as-usual" (BAU) scenario.

• Technological and Policy Adaptations: The report did not fully account for technological advancements, efficiency improvements, and policy interventions that have, in some cases, mitigated resource depletion and environmental damage.

Criticisms and Reassessments

• Overestimation of Resource Limits: Some critics argue that Limits to Growth underestimated humanity’s ability to find substitutes for scarce resources (e.g., renewable energy replacing fossil fuels).

• Underestimation of Innovation: Advances in technology, agriculture, and energy efficiency have delayed some of the dire outcomes.

• Ignored Economic and Market Dynamics: Market adaptations, such as price-driven resource conservation and recycling, were not central in the original model.

Recent Reassessments

Several follow-up studies (e.g., by researchers at MIT and the University of Melbourne) suggest that real-world data since 1972 closely aligns with some of the Limits to Growth scenarios, particularly those predicting economic and environmental strain by the mid-21st century if no significant corrective actions are taken.

Conclusion

While Limits to Growth did not make precise, time-bound predictions, its core message about the risks of unchecked growth in a finite world remains relevant. Some aspects were too pessimistic, but overall, its warning about ecological and economic sustainability has proven insightful.

___________________________________________________________________