What an Artificial Intelligence (AI) Thinks About The Rise and Implications of Artificial General Intelligence (AGI)

What an Artificial Intelligence (AI) Thinks About The Rise and Implications of Artificial General Intelligence (AGI)

In a collaboration between Fast Future, London Futurists, and the UK node of the Millennium Project, we have launched a global discovery programme on AGI, starting with a survey on how it might evolve and impact society, the economy, business, and government. You can take the survey here https://www.surveymonkey.co.uk/r/FFAGI . It takes 7-10 minutes to complete the multiple choice questions, a little more if you want to add comments. Everyone who completes the survey will get a copy of the results and a free eBook from the Fast Future bookstore.

One of our friends from the futurist community Tony Czarnecki (TC) of Sustensis decided to see how Open-AI’s Generative Pre-trained Transformer 3 (GPT-3) natural language processor would answer the questions we raised. The answers presented below are a fascinating insight into the current state of AI capabilities. As you can see, there are a lot of sound and insightful answers, there is also a lot of repetition – evidence, as Tony points out, that we don’t yet have AGI and that GPT-3 isn’t developing a bigger picture awareness of its responses and answering new questions in a different manner as a human might.

To learn more about how Tony conducted this dialogue, here are his details:

Tony Czarnecki - Managing Partner - Sustensis - www.sustensis.co.uk [email protected]

?An interview with a GPT-3 (an AI Natural Language Processor) answering questions to the questionnaire created by Fast Futures and London Futurists.

Interviewer: Tony Czarnecki (TC)

Introduction: This interview follows the original questionnaire created by Fast Future and London Futurists. Generative Pre-trained Transformer 3 (GPT-3) is one of the most advanced Natural Language Processing (NLP) platforms available. The application is supported by 175 billion tokens (machine learning parameters) and was created by Open-AI in July 2020. The GPT-3 database includes all of the entries on Wikipedia, millions of articles, and content from reliable sources of information from across the entire spectrum of knowledge domains. To use it, one must adjust the GPT-3 to your particular needs. One requirement is the creation of a few sentences of context (the primer). Others include setting several parameters, which affect the quality of the answers.

This interview follows all the original questions in the survey, apart from question one, which related to answering personal data. All other questions have been preserved as much as possible. However, from question 8 onwards, the options listed have been substituted by additional sub-questions as necessary annotated by the initials TC.

1. Primer: (this is the initial context for GPT-3)

TC: The world has started to change in most areas at nearly an exponential pace. It is the technology, which is the driving force behind that pace of change, and in particular the development of the AI’s capabilities. AI has already been delivering many benefits to us all and in the next few decades it may create the world of unimaginable wealth.

2. What is the earliest date by which you think there’s a 50% chance that AGI will have been created?

AI Agent: I believe that there is a 50% chance that AGI will have been created by 2040.

TC: Why do you believe so?

AI Agent: The exponential pace of technological change is driven by the development of AI’s capabilities. AI has already been delivering many benefits to us all and in the next few decades it may create the world of unimaginable wealth.

3. Which factors will be the most important in achieving AGI?

AI Agent: The most important factors in achieving AGI will be the development of Artificial General Intelligence and the ability of AI to learn and evolve on its own.

4. What kind of organisation is the most likely to create the first AGI?

AI Agent: The most likely organisation to create the first AGI is a large tech company such as Google, IBM, Microsoft, or Amazon.

TC: Why do you think so?

AI Agent:?Large tech companies have the resources to invest in the development of AI and they also have the expertise to do so. They are also likely to be more open to taking risks in order to achieve breakthroughs.

5. Which factors could have the greatest influence on who might be the first to market with AGI:

AI Agent:?The factors that could have the greatest influence on who might be the first to market with AGI include the level of investment in AI development, the rate of progress in AI development and the level of risk-taking.

6. In which part of the world is AGI most likely to be first created?

AI Agent: AGI is most likely to be first created in the United States, followed by China and Europe.

TC: Why do you think so?

AI Agent: The United States is the most developed country in terms of its technology sector and it has the largest number of tech companies. China is the largest country in terms of population and it has a large number of tech companies. Europe has a large number of tech companies and is also a leading centre of research in AI.

7. Which factors could have the greatest influence on where AGI might be created first?

AI Agent: The factors that could have the greatest influence on where AGI might be created first include the level of investment in AI development, the rate of progress in AI development and the level of risk-taking. (TC comment: It is a similar question to no.3, so there is a similar answer)

8. Which are the most realistic and desirable outcomes of AGI:

AI Agent: Solution to healthcare problems such as cancer and dementia, solutions to climate change, environmental, and sustainability challenges, discovery of new fields of science and technology, Development of new industries, professions, and jobs, Creation of art, literature, and music with unprecedented quality, Creation of fully immersive and personalised AR / VR worlds and experiences, Generation of a superabundance of low-cost clean water, food, energy, etc.

TC: But which one is the most important and why?

AI Agent: I believe that the most important outcome of AGI is the development of new industries, professions, and jobs. This is because it will bring about immense change in all aspects of society, from healthcare to the environment. It will also allow us to explore new areas of science and technology that we have not yet imagined.

TC: Which will be the next most important?

AI Agent: I believe that the next most important outcome of AGI is the generation of a superabundance of low-cost clean water, food, energy, etc. This is because it will solve many of the world's most pressing problems, such as famine and energy shortages. It will also help us to live more sustainably and reduce our impact on the environment.

TC: And what would be the most important after that?

AI Agent: I believe that the most important outcome of AGI after that would be the development of a just and fair society. This is because it would ensure that everyone has the opportunity to benefit from the advances that AGI will bring about. It would also help to reduce inequality and poverty.

9. What you think will be the factors that will determine where the greatest investment will be made in developing AGI solutions?

AI Agent: The factors that will determine where the greatest investment will be made in developing AGI solutions are:

1.The level of investment that has been made in developing AGI solutions to date.

2. The current state of the AGI technology.

3. The potential benefits of AGI solutions.

4. The potential risks of AGI solutions.

10. Which outcomes represent the most serious threats from AGI?

AI Agent: The most serious threats from AGI are:

1. The development of AGI technology that can be used to harm humans or animals.

2. The development of AGI technology that can be used to take over the world.

3. The development of AGI technology that can be used to create artificial intelligence that is smarter than humans.

11. What types of undesirable or unintended consequences could emerge from the rise of AGI?

AI Agent: Some of the undesirable or unintended consequences that could emerge from the rise of AGI include:

1. The development of AGI technology that can be used to harm humans or animals.

2. The development of AGI technology that can be used to take over the world.

3. The development of AGI technology that can be used to create artificial intelligence that is smarter than humans.

4. The development of AGI technology that can be used to automate large numbers of jobs, leading to mass unemployment.

5. The development of AGI technology that can be used to manipulate or control humans. (TC: some repetitions here from the questions above)

12. What are the most likely economic outcomes of AGI?

AI Agent: The most likely economic outcomes of AGI include:

1. The development of AGI technology that can be used to automate large numbers of jobs, leading to mass unemployment.

2. The development of AGI technology that can be used to create artificial intelligence that is smarter than humans.

3. The development of AGI technology that can be used to improve economic efficiency and productivity.

4. The development of AGI technology that can be used to improve human cognitive abilities. (TC: some repetitions here from the questions above)

13. What are the possible economic and financial consequences that could emerge from the rise of AG?

AI Agent: Some of the possible economic and financial consequences that could emerge from the rise of AGI include:

1. The development of AGI technology that can be used to automate large numbers of jobs, leading to mass unemployment.

2. The development of AGI technology that can be used to create artificial intelligence that is smarter than humans.

3. The development of AGI technology that can be used to improve economic efficiency and productivity.

4. The development of AGI technology that can be used to improve human cognitive abilities.

5. The development of AGI technology that can be used to manipulate or control humans. (TC: Again, some repetitions here from the questions above, although the questions are somewhat similar)

14. What are the most likely outcomes for individuals and society when AGI arrives?

AI Agent: The most likely outcomes for individuals and society when AGI arrives are positive, negative, and neutral.

TC: But which are the most likely outcomes?

AI Agent:

Positive:

1. Enhanced knowledge and understanding.

2. Greater efficiency and productivity.

3. Increased ability to solve complex problems.

4. More robust and resilient infrastructure.

5. Greater creativity and innovation.

6. Increased opportunities for learning and growth.

7. More effective and efficient communication.

8. Greater social and economic equity.

9. Enhanced quality of life.

Negative:

1. Increased vulnerability to terrorist attacks.

2. Augmented cyber-security threats.

3. Dwindling natural resources.

4. Greater social and economic inequality.

5. Accelerated environmental degradation.

6. Unrestrained proliferation of weapons of mass destruction.

7. Widespread joblessness and social dislocation.

8. Robust and pervasive surveillance.

9. Authoritarianism and totalitarianism.

Neutral:

1. No significant change.

2. Mostly positive effects, but some negative effects as well.

3. Mostly negative effects, but some positive effects as well.

4. Completely positive or negative effects, with no neutral effects.

15. How individuals and society might approach and prepare for AGI?

AI Agent: There is no one answer to this question, as there is no one "correct" way to approach and prepare for AGI. Some possible approaches include:

1. Focusing on research and development to achieve AGI as quickly as possible.

2. Studying the potential implications of AGI and planning for various possible outcomes.

3. Working to create ethical guidelines for AGI development and use.

4. Preparing for mass unemployment and social dislocation caused by widespread joblessness.

5. Planning for ways to use AGI to improve quality of life and address global challenges.

16. What are the most likely outcomes for governments when AGI arrives?

AI Agent: The most likely outcomes for governments when AGI arrives are positive, negative, and neutral.

TC: But which are the most likely outcomes? (TC: the Agent has to be prompted to expand its answer)

AI Agent:

Positive:

1. Increased ability to solve complex problems.

2. More robust and resilient infrastructure.

3. Greater creativity and innovation.

4. Increased opportunities for learning and growth.

5. More effective and efficient communication.

6. Greater social and economic equity.

7. Enhanced quality of life.

Negative:

1. Increased vulnerability to terrorist attacks.

2. Augmented cyber-security threats.

3. Dwindling natural resources.

4. Greater social and economic inequality.

5. Accelerated environmental degradation.

6. Unrestrained proliferation of weapons of mass destruction.

7. Widespread joblessness and social dislocation.

8. Robust and pervasive surveillance.

9. Authoritarianism and totalitarianism.

Neutral:

1. No significant change.

2. Mostly positive effects, but some negative effects as well.

3. Mostly negative effects, but some positive effects as well.

4. Completely positive or negative effects, with no neutral effects.

17. How national governments and global institutions might approach and prepare for AGI?

AI Agent: There is no one answer to this question, as there is no one "correct" way to approach and prepare for AGI. Some possible approaches include:

1. Focusing on research and development to achieve AGI as quickly as possible.

2. Studying the potential implications of AGI and planning for various possible outcomes.

3. Working to create ethical guidelines for AGI development and use.

4. Preparing for mass unemployment and social dislocation caused by widespread joblessness.

5. Planning for ways to use AGI to improve quality of life and address global challenges.

18. What are the most likely outcomes for businesses when AGI arrives?

AI Agent: The most likely outcomes for businesses when AGI arrives are positive, negative, and neutral.

TC: But which are the most likely outcomes?

AI Agent:

Positive:

1. Enhanced efficiency and productivity.

2. Increased ability to solve complex problems.

3. More robust and resilient infrastructure.

4. Greater creativity and innovation.

5. Increased opportunities for learning and growth.

6. More effective and efficient communication.

7. Greater social and economic equity.

8. Enhanced quality of life.

Negative:

1. Increased vulnerability to terrorist attacks.

2. augmented cyber-security threats.

3. Dwindling natural resources.

4. Greater social and economic inequality.

5. Accelerated environmental degradation.

6. Unrestrained proliferation of weapons of mass destruction.

7. Widespread joblessness and social dislocation.

8. Robust and pervasive surveillance.

9. Authoritarianism and totalitarianism.

Neutral:

1. No significant change.

2. Mostly positive effects, but some negative effects as well.

3. Mostly negative effects, but some positive effects as well.

4. Completely positive or negative effects, with no neutral effects.

19. How businesses might approach and prepare for AGI?

AI Agent: There are many ways businesses might approach and prepare for AGI. One way would be to develop specific strategies for how to best manage AI and automation in order to maintain a competitive advantage. Additionally, businesses could invest in research and development to better understand and prepare for the potential implications of AGI.

TC: What more?

AI Agent: There are many other ways businesses might approach and prepare for AGI. For example, businesses could:

1. Develop policies and procedures for managing AI and automation

2. Create a task force or working group to study and prepare for AGI

3. Partner with universities and other research institutions to study AGI and its potential implications

4. Monitor and track developments in AI and related technologies

TC: Could you think about something else yet?

AI Agent: Yes, businesses might also consider forming a coalition or alliance to discuss and prepare for AGI. Additionally, businesses could invest in public awareness campaigns to educate people about AGI and its potential implications.

20. What skills should be prioritised, in anticipation of the rise of AGI?

AI Agent: Some of the skills that might be prioritised in anticipation of the rise of AGI include:

• critical thinking
• problem solving
• creativity
• adaptability
• collaboration
• leadership

TC: Anything more?

AI Agent: There are many other skills that could be prioritised in anticipation of the rise of AGI. For example, businesses might focus on developing employees' technology skills, as well as their skills in data analysis and interpretation. Additionally, businesses might emphasise the importance of teamwork and collaboration, as these skills will be essential in a world where AI and automation play a larger role.

21. What are the skills challenges presented by AGI?

AI Agent: The skills challenges presented by AGI are largely similar to the challenges presented by AI and automation. In a world where AI and automation are prevalent, businesses will need employees who are skilled in critical thinking, problem solving, and creativity. Additionally, businesses will need employees who are adaptable and able to work in a changing environment.

TC: But which is the most significant challenges that AGI presents for new skills?

AI Agent: The most significant challenge that AGI presents for new skills is its potential to replace human workers. With AGI, businesses will be able to automate many tasks that were once performed by humans. This could result in a workforce that is less skilled and less adaptable, which could have a negative impact on businesses.

22. Which are the best approaches to governing the rise of AGI?

AI Agent: There is no one-size-fits-all answer to this question, as the best approach to governing the rise of AGI will vary from country to country and even from region to region. However, some of the approaches that might be considered include:

1. Develop policies and procedures for managing AI and automation

2. Create a task force or working group to study and prepare for AGI

3. Partner with universities and other research institutions to study AGI and its potential implications

4. Monitor and track developments in AI and related technologies

TC: Anything more?

AI Agent: There are many other approaches that could be used to govern the rise of AGI. For example, businesses and governments could work together to create an international strategy for AGI. Additionally, governments could invest in research and development to better understand AGI and its potential implications.

23. Would you like to share any further comments on the governance of AGI?

AI Agent: The governance of AGI is a difficult problem that has not been solved yet. One issue is how to ensure that AGI is safe and does what we want it to do. Another issue is how to prevent AGI from being used for malicious purposes.

TC: But what would be your recommendation?

AI Agent: I think it is important to have a good understanding of the risks and benefits of AGI before deploying it. We need to make sure that AGI is safe and does what we want it to do. We also need to make sure that it is not used for malicious purposes.

TC: Anything else?

AI Agent: I think it is important to research and develop AGI in a responsible way. We need to make sure that AGI is safe and does what we want it to do. We also need to make sure that it is not used for malicious purposes.

TC: But regarding AGI governance which is your favourite approach?

AI Agent: I think it is important to have a good understanding of the risks and benefits of AGI before deploying it. We need to make sure that AGI is safe and does what we want it to do. We also need to make sure that it is not used for malicious purposes.

TC: I understand. However, we have very little time to control AI before it becomes AGI. Therefore we need concrete solutions? Which would your most preferred way forward to govern AGI?

AI Agent: I think it is important to have a good understanding of the risks and benefits of AGI before deploying it. We need to make sure that AGI is safe and does what we want it to do. We also need to make sure that it is not used for malicious purposes. (TC: some repetitions in the final answer, which means we do not have AGI yet!)

You can take the survey here https://www.surveymonkey.co.uk/r/FFAGI .