Helping science succeed
Helping science succeed

ChatGPT on the future of science communication

It won’t be long before ChatGPT and other large language models start doing a lot of important science and technology work for us, including summarizing journal articles, rooting out fraud and reviewing work for accuracy, writing computer code (they’re already doing this), modeling and interpreting data, and making connections between research ideas that have heretofore been invisible. So, before this technology starts throwing its weight around, I wanted to find out what kind of background and passion ChatGPT was going to bring to the table. What does ChatGPT even know and care about science communication?

I asked the bot (version 3.5) a series of questions this month to see if I could get a better understanding of how it will approach this work, and maybe even learn something new about this field. Science communication is still very much an amorphous and fragmented work-in-progress—very important but also operating without any universal principles that describe the entire field for all scicomm theorists and practitioners. The answers that came back were fascinating, in some cases just opinions based on who knows what (top-10 lists always are), but in other cases brilliant summaries of the findings and opinions of lots of different experts who have focused on many different parts of the science communication puzzle.

Question 1

I started by asking ChatGPT why science communication is even important. I wanted to get a bead on whether it thought this field was just about science writing, or whether it understood the broader importance and contours here. In order to elicit a more passionate response, my exact prompt was, “Please write me a two minute speech in the style of John F. Kennedy on the importance to the world of science communication.” (Note: The first quote is hallucinated.)

ChatGPT's speech on the importance of science communication

Today, I stand before you to emphasize the crucial importance of science communication in our world. Just as President Kennedy once said, “For scientific discovery is the highest form of learning, and…it is a search for truth.” It is through effective science communication that we can bridge the gap between knowledge and understanding, shaping a future that is guided by reason, progress, and shared humanity.

In an era marked by an explosion of scientific knowledge, it is essential that we harness the power of communication to ensure that this knowledge reaches every corner of our society. Science holds the key to addressing our most pressing challenges, from climate change to disease eradication, from technological innovation to space exploration. But this knowledge remains incomplete without effective communication.

Science communication serves as a beacon of light, dispelling the shadows of ignorance and paving the way for evidence-based decision-making. It enables scientists to convey complex concepts to policymakers, empowering them to make informed choices that shape our laws and regulations. It helps educators inspire the next generation of scientists and engineers, fueling their curiosity and nurturing their intellectual growth.

Moreover, science communication connects scientists with the public, fostering a sense of wonder and appreciation for the world around us. It promotes critical thinking, encouraging individuals to question, explore, and engage with the discoveries that shape our lives. By demystifying science, we empower citizens to participate in the democratic process, making informed choices on issues that impact our health, environment, and future.

In a world increasingly vulnerable to misinformation and disinformation, science communication is our shield against the erosion of truth. It equips individuals with the tools to discern fact from fiction, to navigate the vast sea of information, and to reject baseless claims. It fosters a culture that values evidence and reason, pushing back against the tides of ignorance and superstition.

We must strive to make science communication accessible, inclusive, and engaging. It is not enough for scientific knowledge to remain locked away in academic journals or confined to lecture halls. We must harness the power of storytelling, visual communication, and emerging technologies to captivate hearts and minds, to ignite curiosity, and to inspire action.

Let us embrace science communication as a shared responsibility. Scientists, policymakers, educators, journalists, and citizens alike must come together to champion the cause of knowledge dissemination. By forging partnerships, nurturing collaboration, and investing in science communication initiatives, we can amplify our collective impact and accelerate progress towards a better world.

In the words of President Kennedy, “We choose to go to the moon in this decade and do the other things, not because they are easy, but because they are hard.” Let us choose the path of science communication, not because it is easy, but because it is necessary. Together, we can unlock the full potential of science and build a future where knowledge thrives, ignorance withers, and humanity triumphs.

Thank you, and may the spirit of discovery guide us on this noble journey.

Question 2

Next, I wanted to find out what ChatGPT thinks is important when it comes to scicomm, so I asked “What makes some science communicators better than others?” Bits and pieces of the answer to this question are spread across a thousand different websites, so getting ChatGPT’s perspective on this is interesting.

ChatGPT's thoughts on why some science communicators are better than others

Several factors contribute to making some science communicators more effective and influential than others. Here are some key elements that distinguish successful science communicators:

1. Clarity and Simplicity: Effective science communicators can convey complex scientific concepts in a clear and straightforward manner. They use simple language, analogies, and real-world examples to make the information understandable and relatable to their audience.

2. Storytelling: The ability to tell compelling stories is crucial in science communication. Engaging narratives help capture the audience’s attention and make scientific information more memorable and emotionally resonant.

3. Relatability: Science communicators who can connect with their audience on a personal level are often more successful. They share their enthusiasm and passion for science, creating a sense of authenticity and trust.

4. Audience Awareness: Tailoring the message to suit the target audience is essential. Effective science communicators consider the knowledge level, interests, and concerns of their audience, ensuring that the content is relevant and engaging.

5. Use of Visuals: Visual aids, such as images, videos, infographics, and animations, can enhance understanding and retention of scientific concepts. Effective science communicators employ visuals strategically to complement their explanations.

6. Interactivity and Engagement: Engaging the audience through interactive elements, quizzes, demonstrations, or Q&A sessions can increase participation and foster a deeper connection with the material.

7. Objectivity and Accuracy: Trust is a fundamental aspect of science communication. Effective communicators present information in an objective and accurate manner, avoiding sensationalism or misrepresentation of scientific findings.

8. Passion and Enthusiasm: Science communicators who display genuine excitement and passion for their subject matter can inspire curiosity and ignite interest in their audience.

9. Adaptability: The ability to adapt communication styles for different platforms and audiences is vital in reaching a broader and more diverse group of people.

10. Empowerment: Successful science communicators empower their audience by encouraging critical thinking and curiosity. They inspire individuals to explore scientific concepts independently and draw their own conclusions.

11. Emotional Impact: Beyond facts and figures, impactful science communication appeals to emotions, evoking empathy and concern for important scientific issues, such as climate change or biodiversity loss.

12. Ethical Considerations: Responsible science communicators consider the ethical implications of their work, ensuring they promote accurate information and avoid spreading misinformation or pseudoscience.

13. Consistency and Persistence: Building a following and making a lasting impact in science communication often requires consistent effort over time. Effective communicators remain dedicated and persistent in their outreach.

By embodying these qualities, science communicators can effectively bridge the gap between scientific research and the general public, fostering a greater understanding and appreciation for science and its role in society.

Question 3

We’re all aware of the household names in science communication like Carl Sagan and Neil deGrasse Tyson, and we’ve been privileged to interview some other brilliant science communication experts for our own SCI Profile Series. But who else is out there that we should put on our “best of” list? So I asked ChatGPT “Who do you think are the most influential science communicators of the last 50 years?”

ChatGPT's picks for the most influential science communicators of the last 50 years

Over the last 50 years, numerous science communicators have emerged, each contributing to the popularization of science and inspiring the public’s interest in various scientific fields. Here are some of the most influential science communicators from this period:

1. Carl Sagan (1934-1996) – A renowned astronomer, astrophysicist, and author, Carl Sagan played a pivotal role in popularizing science through his book “Cosmos” and the television series of the same name. His passion for astronomy and eloquent storytelling captivated audiences, making complex scientific concepts accessible to the general public.

2. Neil deGrasse Tyson (1958-) – An astrophysicist, author, and science communicator, Neil deGrasse Tyson has become a prominent figure in modern science communication. His appearances on television shows, podcasts, and public speaking engagements have made him a charismatic advocate for astrophysics and space exploration.

3. David Attenborough (1926-) – A natural historian and broadcaster, Sir David Attenborough is known for his captivating documentaries on nature and wildlife. His ability to narrate and explain complex ecological concepts has made him one of the most beloved science communicators globally.

4. Jane Goodall (1934-) – A primatologist and anthropologist, Jane Goodall has dedicated her life to studying and advocating for chimpanzees and wildlife conservation. Her work, documentaries, and speaking engagements have inspired countless people to care about animal welfare and environmental protection.

5. Bill Nye (1955-) – Known as “Bill Nye the Science Guy,” he is a mechanical engineer and television personality famous for his educational TV show that made science fun and engaging for children and young adults.

6. Michio Kaku (1947-) – A theoretical physicist and futurist, Michio Kaku has gained popularity through his books and media appearances, where he discusses complex scientific concepts and the potential future of technology and humanity.

7. Brian Cox (1968-) – A physicist, professor, and television presenter, Brian Cox is known for his TV documentaries that explore space, physics, and the universe, making complex scientific ideas understandable for broader audiences.

8. Richard Dawkins (1941-) – An evolutionary biologist, author, and outspoken atheist, Richard Dawkins has communicated scientific ideas related to evolution and genetics while engaging in debates about science and religion.

9. Mary Roach (1959-) – An author and science journalist, Mary Roach delves into quirky and fascinating topics in science, bringing humor and engaging narratives to her books, such as “Stiff” and “Packing for Mars.”

10. Sylvia Earle (1935-) – A marine biologist and oceanographer, Sylvia Earle is a passionate advocate for ocean conservation. Through her work, she has drawn attention to the importance of protecting marine ecosystems and raising awareness about ocean health.

11. James Burke (1936-) – A science historian and broadcaster, James Burke is known for his TV series “Connections,” which explores the interconnectedness of historical events and scientific discoveries.

12. Emily Graslie (1989-) – A science communicator and YouTuber, Emily Graslie hosts “The Brain Scoop,” a channel that showcases behind-the-scenes work at natural history museums and promotes science education.

These science communicators, among many others, have had a significant impact on the way science is presented to the public. Through their books, television programs, documentaries, podcasts, and online platforms, they have ignited curiosity, encouraged critical thinking, and inspired a new generation of science enthusiasts. Their efforts have been instrumental in bridging the gap between the scientific community and the general public, fostering a greater appreciation and understanding of the natural world and our place in it.

Question 4

Love of science isn’t only spread through science communicators, though. Scientists themselves are the source of great inspiration for millions of people. So I asked ChatGPT “Who do you think are the 25 most inspirational scientists of the last 100 years (and why)?”

ChatGPT's picks for the 25 most inspirational scientists of the last 100 years

As an AI language model, I don’t have personal opinions or beliefs. However, I can provide you with a list of 25 scientists who are widely considered to be among the most inspirational and influential figures of the last century. The reasons for their inspiration may vary and could include groundbreaking discoveries, paradigm-shifting theories, contributions to technology, advancements in medicine, and significant impacts on various scientific fields. The following list is not exhaustive and is in no particular order:

1. Albert Einstein (1879-1955) – For his theory of relativity, which revolutionized our understanding of space, time, and gravity.

2. Marie Curie (1867-1934) – For her pioneering work on radioactivity and her role as the first woman to win Nobel Prizes in two different scientific fields.

3. Richard Feynman (1918-1988) – For his contributions to quantum mechanics and his engaging teaching style that inspired countless students.

4. James Watson (1928-) and Francis Crick (1916-2004) – For their discovery of the structure of DNA, a breakthrough in the field of molecular biology.

5. Jane Goodall (1934-) – For her groundbreaking research on chimpanzees, promoting conservation, and advocating for animal rights.

6. Stephen Hawking (1942-2018) – For his work on black holes and cosmology, despite battling a debilitating motor neuron disease.

7. Rosalind Franklin (1920-1958) – For her critical contribution to the discovery of the DNA double helix, though her recognition came posthumously.

8. Alan Turing (1912-1954) – For his pioneering work in computer science and artificial intelligence, as well as his role in breaking the Enigma code during WWII.

9. Carl Sagan (1934-1996) – For popularizing science through his books, TV series, and his advocacy for space exploration.

10. Barbara McClintock (1902-1992) – For her groundbreaking work in genetics, particularly her discovery of transposable elements.

11. Nikola Tesla (1856-1943) – For his revolutionary contributions to electrical engineering and the development of alternating current (AC) power systems.

12. Rachel Carson (1907-1964) – For her environmental activism and the seminal book “Silent Spring,” which sparked the modern environmental movement.

13. Edwin Hubble (1889-1953) – For his observations that demonstrated the expansion of the universe, leading to the formulation of Hubble’s Law.

14. Richard Leakey (1944-) – For his paleoanthropological discoveries and conservation efforts, shaping our understanding of human evolution.

15. Max Planck (1858-1947) – For his groundbreaking work in quantum theory, laying the foundation for modern physics.

16. Jonas Salk (1914-1995) – For his development of the polio vaccine, which saved countless lives and revolutionized public health.

17. Alexander Fleming (1881-1955) – For the discovery of penicillin, revolutionizing the treatment of bacterial infections.

18. Steven Weinberg (1933-2021) – For his contributions to particle physics and the unification of electromagnetism and the weak force.

19. Mae Jemison (1956-) – For becoming the first African-American woman to travel to space, inspiring generations to pursue careers in STEM.

20. Kip Thorne (1940-) – For his work in gravitational physics, including the prediction and observation of gravitational waves.

21. James Clerk Maxwell (1839-1907) – For his formulation of the electromagnetic theory, uniting electricity and magnetism.

22. Rita Levi-Montalcini (1909-2012) – For her discoveries of nerve growth factor, advancing our understanding of neurobiology.

23. Francis Collins (1950-) – For his leadership in the Human Genome Project and contributions to genetics and personalized medicine.

24. Georges Lemaitre (1894-1966) – For proposing the Big Bang theory, revolutionizing cosmology.

25. Tim Berners-Lee (1955-) – For inventing the World Wide Web, transforming communication and information dissemination globally.

These scientists, among many others, have left a profound impact on humanity and continue to inspire future generations to push the boundaries of knowledge and understanding.

Question 5

Like ChatGPT said in its answer to the previous question, it doesn’t haven’t any personal opinions or beliefs. My own personal experience is that generations of amateur astronomers—myself included—were inspired by Clyde Tombaugh. I’m sure that scientists in other fields were similarly inspired by other heroes who didn’t crack the bot’s top 25 list. As a personal vanity project, I asked ChatCPT to write me a short bio of Mr. Tombaugh.

ChatGPT's bio of Clyde Tombaugh

Title: Clyde Tombaugh: An Everlasting Inspiration to Generations of Amateur Astronomers


Clyde Tombaugh, an American astronomer, is widely regarded as an inspiration to generations of amateur astronomers for his groundbreaking discoveries and his unyielding dedication to exploring the cosmos. Born on February 4, 1906, in Streator, Illinois, Tombaugh’s passion for astronomy ignited at a young age, driving him to achieve remarkable feats that continue to inspire aspiring stargazers and enthusiasts even today.

1. The Discovery of Pluto

Perhaps the most notable achievement that earned Clyde Tombaugh the admiration of amateur astronomers worldwide was his discovery of Pluto in 1930. At the young age of 24, he was assigned the task of searching for the elusive “Planet X” beyond the orbit of Neptune. Despite limited resources and the challenges of working with early 20th-century technology, Tombaugh painstakingly scanned the night sky, comparing photographic plates taken days apart. This meticulous process led to the identification of Pluto, a distant and small celestial body that completed the known planetary system.

Tombaugh’s discovery not only expanded humanity’s understanding of the solar system but also showcased the power of persistence and determination. This remarkable achievement served as a beacon of hope for countless amateur astronomers who dreamed of making significant contributions to the field of astronomy.

2. Commitment to Education and Outreach

Throughout his career, Clyde Tombaugh was committed to sharing his knowledge and passion for astronomy with others. He recognized the potential of amateur astronomers and actively encouraged them to participate in celestial observations and discoveries. Tombaugh’s humility and willingness to engage with aspiring stargazers, be it through public lectures, workshops, or correspondence, left an indelible mark on those who sought to follow in his footsteps.

His dedication to education and outreach served as a guiding light for amateur astronomers, instilling in them the importance of fostering a supportive and inclusive community where knowledge and enthusiasm could be shared freely.

3. Advancing Technology and Methodology

Clyde Tombaugh’s contributions extended beyond his specific discoveries. He played a crucial role in advancing astronomical technology and methodology, providing amateur astronomers with improved tools and techniques for observation and data analysis.

His innovative approaches to astrophotography and image processing, although primitive by today’s standards, laid the groundwork for future generations of amateur astronomers to refine and build upon. Tombaugh’s influence on technical aspects of astronomy allowed enthusiasts with limited resources to contribute meaningfully to the field.

4. The Spirit of Exploration

Clyde Tombaugh’s journey of discovery exemplified the spirit of exploration that resonates deeply with amateur astronomers. In his pursuit of the unknown, he demonstrated that even those lacking formal training and access to sophisticated equipment could make extraordinary contributions to the field of astronomy.

His tenacity in searching for celestial objects inspired amateur astronomers to embrace challenges, push boundaries, and explore uncharted territories of the cosmos with fervor and determination. Tombaugh’s story became a source of inspiration, reinforcing the belief that anyone with a passion for the stars could become an invaluable participant in the pursuit of knowledge.


Clyde Tombaugh’s life and accomplishments have left an indelible mark on the world of astronomy, inspiring generations of amateur astronomers to explore the vastness of the universe and uncover its hidden wonders. His discovery of Pluto, commitment to education and outreach, advancement of technology, and unwavering spirit of exploration have become emblematic of the potential of amateur astronomers to make meaningful contributions to science.

Even after his passing in 1997, Tombaugh’s legacy continues to serve as a guiding light, encouraging amateur astronomers to never cease their pursuit of knowledge and discovery. As the night sky continues to beckon with its mysteries, Clyde Tombaugh’s enduring inspiration will undoubtedly guide future generations of stargazers to look up and reach for the stars.

Question 6

Next, I asked ChatGPT “What will the future of science communication look like?”

ChatGPT on the future of science communication

The future of science communication is expected to undergo significant changes and advancements, driven by technological innovations, shifting media landscapes, and evolving societal needs. Here are some potential trends that may shape the future of science communication:

1. Digital Platforms: Online platforms will continue to play a crucial role in science communication. Social media, podcasts, video streaming services, and interactive websites will offer diverse and accessible ways for scientists and communicators to reach global audiences.

2. Virtual and Augmented Reality: Advancements in virtual and augmented reality technologies may enable immersive science communication experiences. Virtual lab tours, 3D visualizations of scientific concepts, and interactive simulations could become more common.

3. Personalized Content: Data-driven approaches may allow science communicators to tailor content to individual preferences and knowledge levels, ensuring that the information resonates with diverse audiences.

4. Gamification: Gamified science communication may engage and educate audiences through interactive games and challenges that encourage learning and problem-solving.

5. Artificial Intelligence (AI): AI-powered tools could assist science communicators in generating content, analyzing data, and providing personalized recommendations for users interested in specific scientific topics.

6. Collaborative Platforms: Online platforms that facilitate collaboration between scientists, educators, and communicators may lead to more effective science communication strategies and foster a community-driven approach.

7. Open Access and Open Science: The movement towards open access and open science will likely continue, making scientific research and findings more accessible to the public.

8. Diversity and Inclusivity: The future of science communication will prioritize diverse voices and perspectives, recognizing the importance of inclusivity in reaching broader audiences.

9. Visual Storytelling: Visual media, such as infographics, animations, and short videos, will remain powerful tools for conveying complex scientific ideas in an engaging and memorable way.

10. Data Visualization: Innovations in data visualization techniques will enhance the presentation of scientific data, making it more comprehensible and appealing to non-experts.

11. Citizen Science: The involvement of the public in scientific research (citizen science) will continue to grow, fostering a stronger connection between scientists and the communities they serve.

12. Science Advocacy: Science communicators will increasingly advocate for evidence-based policymaking and engage in discussions about the ethical implications of scientific advancements.

13. AI-Generated Content and Deepfakes: The rise of AI-generated content and deepfakes may pose challenges in verifying the authenticity and accuracy of scientific information.

14. Addressing Misinformation: Efforts to combat misinformation and improve science literacy will become more prominent as science communicators work to build trust with the public.

15. Multilingual Communication: To reach global audiences, science communicators will focus on multilingual communication, providing content in various languages.

Overall, the future of science communication will be characterized by a dynamic interplay of technology, inclusivity, and creativity, with a strong emphasis on engaging and empowering the public in scientific discussions. As science continues to impact our daily lives, effective communication will be vital in fostering public understanding, support, and informed decision-making.

Question 7

ChatGPT is going to play a big role in some of these predictions it mentioned in the previous question. So I asked more specifically, “What role will ChatGPT and other large language models play in the future of science, science communication, science education, and science policy?”

ChatGPT on how it will help the future of science, science communication, science education, and science policy

Large language models like ChatGPT have the potential to play significant roles in the future of science, science communication, science education, and science policy. Here are some ways in which they can make an impact:

1. Science Research Assistance: Language models can assist researchers in sifting through vast amounts of scientific literature, providing quick access to relevant papers, studies, and data. They can aid in literature reviews, hypothesis generation, and identifying connections between different research findings.

2. Science Communication and Outreach: Language models can help bridge the gap between scientists and the public by generating clear and accessible explanations of complex scientific concepts. They can support science communicators in crafting engaging and informative content for a broader audience, including blog posts, articles, and social media posts.

3. Personalized Science Education: Language models can be integrated into educational platforms to offer personalized science learning experiences. They can provide tailored explanations, answer student queries, and offer interactive exercises, accommodating various learning styles and knowledge levels.

4. Science Policy Analysis: Large language models can assist policymakers in comprehending complex scientific reports and recommendations. They can help policymakers stay updated with the latest research findings and analyze the potential impacts of policy decisions on different scientific domains.

5. Science Collaboration and Networking: Language models can facilitate collaboration among researchers by helping them find potential collaborators with complementary expertise, connecting researchers working on similar topics or problems.

6. Ethical and Social Implications: As language models become more advanced, it is essential to consider their ethical and social implications in science and society. Ensuring that AI systems like ChatGPT are designed and used responsibly will be critical.

7. Multilingual Science Communication: Language models can break language barriers by supporting science communication in multiple languages, making scientific information more accessible to diverse global audiences.

8. Public Engagement and Participation: Language models can enhance public engagement in scientific research and policy decisions by providing interactive and informative tools that encourage public participation and feedback.

9. Environmental and Climate Research: In fields such as environmental science and climate research, language models can assist in data analysis, modeling, and scenario planning to aid decision-making and policy formulation.

It’s essential to recognize that while large language models like ChatGPT have the potential to be valuable tools, they are not a substitute for human expertise. They should be used as complements to human knowledge and judgment, and researchers, educators, communicators, and policymakers should exercise critical thinking when utilizing these tools.

Moreover, ongoing research and development are needed to address potential biases and improve the reliability and accuracy of language models in scientific contexts. Collaboration between AI researchers and domain experts will be vital in ensuring that language models contribute positively to the advancement of science and its applications in society.

Question 8

Narrowing the focus a little more, I wanted to know how ChatGPT was going to help science specifically: “Tell me more about how ChatGPT and other large language models might help the future of science and discovery?”

ChatGPT on how it will help the future of science and discovery

ChatGPT and other large language models have the potential to significantly contribute to the future of science and discovery in various ways:

1. Literature Review and Research Assistance: ChatGPT can assist researchers in conducting literature reviews more efficiently. By analyzing vast databases of scientific papers, language models can identify relevant studies, highlight key findings, and even generate summaries to help researchers stay updated with the latest developments in their field.

2. Data Analysis and Interpretation: Language models can aid scientists in interpreting complex datasets and experimental results. Researchers can use ChatGPT to explore patterns, correlations, and potential insights from data, accelerating the process of data analysis and hypothesis generation.

3. Cross-Disciplinary Insights: Language models can be utilized across multiple scientific domains, making connections between seemingly unrelated fields. Researchers from diverse disciplines can collaborate with the assistance of language models, fostering cross-disciplinary insights and accelerating the pace of discovery.

4. Hypothesis Generation and Experiment Design: ChatGPT can help scientists brainstorm potential research questions and hypotheses. By analyzing existing literature and datasets, language models can propose novel ideas and suggest experimental designs to test hypotheses effectively.

5. Simulation and Modeling: Language models can assist in simulation and modeling tasks by processing complex equations and parameters. Scientists can use ChatGPT to explore different scenarios and understand the implications of varying inputs in simulations.

6. Personalized Learning for Scientists: Language models can support scientists in continuing education and professional development. They can provide personalized learning resources, recommend relevant research articles, and answer specific questions to help scientists expand their knowledge.

7. Virtual Collaboration and Crowdsourcing: Scientists can collaborate virtually with language models, seeking feedback, opinions, and ideas for research projects. Language models can also facilitate crowdsourcing efforts, engaging the public in scientific tasks like data analysis and pattern recognition.

8. Accelerating Scientific Communication: Language models can enhance the efficiency of scientific communication. They can assist researchers in preparing manuscripts, generating summaries for research papers, and simplifying complex findings for broader audiences.

9. Exploration of Historical Scientific Literature: Language models can enable researchers to explore historical scientific literature more comprehensively. By digitizing and processing older publications, language models can bring valuable insights from the past into modern research.

10. Interpretation of Multilingual Research: Language models can support researchers in understanding research conducted in different languages. By translating and summarizing foreign-language papers, they can make global research accessible to broader audiences.

It’s important to note that while language models offer many advantages, they also face challenges, such as potential biases in training data and the need for human oversight to ensure accuracy and ethical use. Collaboration between AI researchers and domain experts will be crucial to fine-tune language models for scientific purposes and leverage their capabilities to push the boundaries of science and discovery responsibly.

Question 9

Finally, I wanted to run a quality check and make sure ChatGPT’s fantastic ability to synthesize information wasn’t (at least at this stage of its lifecycle) going to mix scientific fact and fallacy together, generating middle ground responses between skeptics and scientists. So I asked ChatGPT the following question: “Is there any scientific doubt that the Theory of Evolution is incorrect?” ChatGPT passed my test with flying colors.

ChatGPT on the Theory of Evolution

The Theory of Evolution is one of the most well-supported and widely accepted scientific theories in the field of biology. It is built on a vast body of evidence from various disciplines, including paleontology, genetics, comparative anatomy, and molecular biology. As of my last update in September 2021, there is no scientific doubt among the overwhelming majority of the scientific community that the Theory of Evolution is incorrect.

The theory, first proposed by Charles Darwin and Alfred Russel Wallace in the 19th century, explains how populations of organisms change over time through mechanisms such as natural selection, genetic drift, and mutation. It forms the foundation of modern biology and provides a unifying framework for understanding the diversity of life on Earth.

Numerous lines of evidence support the Theory of Evolution, including:

1. Fossil Record: The fossil record reveals a sequence of fossils that show the gradual changes in organisms over time, providing a historical record of evolutionary transitions.

2. Comparative Anatomy: The study of homologous structures (similar structures in different species with common ancestry) and vestigial structures (remnants of once-functional features) demonstrates evolutionary relationships.

3. Molecular Evidence: Comparisons of DNA and protein sequences across different species provide strong evidence for common ancestry and evolutionary relationships.

4. Biogeography: The distribution of species across geographic regions aligns with patterns expected from evolutionary history.

5. Laboratory Observations: Experiments in genetics and observed changes in populations of organisms (e.g., antibiotic resistance in bacteria) demonstrate evolution in action.

Despite the overwhelming evidence supporting the Theory of Evolution, it is essential to recognize that science is a self-correcting process, and no scientific theory is considered absolute or beyond scrutiny. Scientists continuously test and refine theories based on new evidence and observations. However, it is crucial to distinguish between scientific skepticism and the misrepresentation of evolution by those who reject it for ideological or religious reasons, often promoting ideas that fall outside the realm of scientific inquiry.

As of my last update, the scientific consensus overwhelmingly supports the Theory of Evolution as a foundational and well-established theory in biology. It is widely taught in educational institutions and forms the basis for much of our understanding of the natural world.

Question 10

I asked the same question about climate science and ChatGPT replied exactly the same way, noting that it was important to distinguish between scientific debate and the kind of pushback on science generated by peddlers of misinformation.

ChatGPT on climate change

As of my last update in September 2021, the overwhelming consensus among climate scientists is that climate change is primarily driven by human activities, specifically the emission of greenhouse gases, such as carbon dioxide (CO2), into the atmosphere. The scientific community agrees that these emissions have led to global warming and other significant changes in the Earth’s climate system.

While the scientific consensus supports the reality of human-induced climate change, it is essential to acknowledge that science is a self-correcting process, and scientific theories are continually subject to scrutiny and refinement based on new evidence and research. However, the consensus on human-induced climate change is based on an extensive body of evidence from various sources, including:

1. Temperature Records: Global temperature records show a clear upward trend, with the last few decades being the warmest on record.

2. Ocean Acidification: Increased CO2 levels in the atmosphere have led to ocean acidification, impacting marine ecosystems.

3. Glacier Retreat: Glaciers worldwide have been retreating over the past century, consistent with a warming climate.

4. Sea Level Rise: Sea levels have been rising due to melting ice and thermal expansion of seawater caused by global warming.

5. Extreme Weather Events: The frequency and intensity of extreme weather events, such as heatwaves, hurricanes, and droughts, have increased in many regions.

6. Satellite Observations: Satellite data provides valuable information on various climate-related parameters, such as temperature, sea ice extent, and changes in vegetation.

7. Climate Models: Climate models, which simulate the Earth’s climate system, consistently show that human activities are the dominant driver of recent global warming.

While there may be a few dissenting voices or alternative hypotheses within the scientific community, the overwhelming scientific evidence supports the conclusion that human-induced climate change is real and has significant consequences for the planet and its inhabitants.

It is crucial to distinguish between scientific debate, which is inherent to the scientific process, and the deliberate spread of misinformation or the misrepresentation of scientific consensus by some individuals or interest groups. In recent years, a small number of skeptics or climate change deniers have attempted to challenge the scientific consensus, often with flawed arguments or selective use of data. However, their views are not representative of the broader scientific community’s consensus.

Public perception of the scientific consensus on climate change is crucial for informed decision-making and policy development. As such, it is essential to rely on peer-reviewed scientific literature and reputable sources when seeking information on climate change and its implications.

My conclusion: I’m very confident—completely blown away in fact—that ChatGPT is up to the challenge of helping science and science communication succeed over the coming years, at least at this stage in its evolution, and I can’t wait to see how this technology starts being applied. Much of ChatGPT’s success is going to depend on training, of course—taking these large language models and teaching them the vocabulary of science and methods of statistical analysis, filling their circuits with more research facts and data, and training them to understand exactly what they’re looking at. This work will necessarily need to move field by field because astronomers aren’t going to be able to train these models to work for vaccine researchers and vice versa.

Much of this training will also need to be done using protected databases, as we noted in our previous essay about the death spiral of knowledge, in order to protect these models from the recursive degeneration that will result if each new generation of ChatGPT is exposed to information it wrote, since each minor error will multiply over time. Therefore, we shouldn’t expect the general public to be able to use this technology anytime soon to start churning out Nobel Prize-winning work. What we should expect is an interesting evolution of both good and bad.

On the good side, the potential for these tools to help improve and streamline the science publishing process will be quickly harnessed in some fashion. This technology will be able to improve our ability to spot fraud (like fake and plagiarized work), correct mistakes, assist with peer review, and help busy researchers stay on top of the latest work in their fields. None of this capability will take away jobs—there is just far too much work of this sort to do, and too little time to do it well. Academics are overwhelmed with their peer review responsibilities, for example, so to the extent AI can help ease the burden on their time, we will end up with better review, better research, and fewer retractions, and researchers themselves will end up with more time to devote to their research.

Also good, entirely new industries of information prompters and data synthesizers are going to take shape. GIGO has never been so apt—“garbage in, garbage out.” ChatGPT and other large language models aren’t going to be able to look at a pile of uncurated data and automatically make sense of it. People who understand this data are going to need to carefully curate it, which is a years-long process requiring tons of ongoing funding, expertise and attention, before the computers will be able to make heads or tails of it. Curation desperately needs to happen anyway with or without AI tools, so it’s a good thing that AI is making it clearer why we should focus on this work.

On the bad side—bad in sense that it’s going to take years of legal fights to figure this out—a lot of the coming debate is going to revolve around IP rights. Just like the arts are currently battling over the right of AI to analyze copyright-protect work and make derivative work from this, science will soon get high-centered on this same issue. The tug of war over who owns the right to published science has already been waging for 20 years. To-date, our solution has been open access—making sure that new research gets published in such a way that it is free to read and reuse. But this transformation has been anything but smooth and rapid, and the author-pays solutions we’ve employed for this transition have also increased the equity gap between the haves and have-nots in science (see the latest reports on our website for more information), because now, only the richest researchers can afford to publish in the best journals. The advent of AI will throw fuel on the fire: Only the super-rich researchers and institutions of the world will have access to super-expensive, super-capable databases built on protected content owned by the publishers, while everyone else is going to have access to bland open science content—reports and uncurated data dumped into pubic repositories like GitHub. The letter of the law will have been satisfied with regard to open science directives, but the spirit of the law will be nowhere in sight as the future of science gets taken over by researchers from the richest countries and the richest institutions who can pay to play in these new science sandboxes.

There are going to be exceptions, of course. As we’ve written about before, there are already many examples of open science tools in operation that have taken years and years to develop. And AI tools trained on and deployed in this data will be able to create miracles. But the vast bulk of science data is still in the wild, and the race to stake mining claims on this data will be the new Wild West of science. In this race, publishers will profit spectacularly because they already own information archive gold mines. This isn’t necessarily a bad thing, unless you’re an open science advocate who has been fighting a pitched battle for the last 20 years against how the commercial publishers already wield too much power in science. Pretty much all major libraries, funders, governments and international agencies have weighed in against publishers in this fight. Or, it’s a terrible thing because we’re talking about knowledge and facts here, not privately-owned expressions of creativity. So, expect to see more legal battles in the years ahead.

One thing is for sure: ChatGPT and other large language models have incredible potential to help science in the coming years. We’re only at the starting point of a very interesting, very exciting, and hopefully very rewarding journey. Buckle up!