Gathered around a conference room table, with one person scribbling on the whiteboard, a group of industry veterans sizzles with energy. Participants build on each other’s thoughts and bounce from one concept to the next, coming up with ideas none of them could have generated individually.
Ah, the group brainstorm. There’s nothing quite like it.
Or is there?
GeekWire is relaunching its “Bot or Not” series today, exploring the line between human ingenuity and artificial intelligence, just as the new era of generative AI makes that line blurrier than ever. This editorial series is reported and written independently by GeekWire’s news team, with underwriting support from Google Cloud. (Read more about underwriting and sponsorship on GeekWire.)
We plan to interview artificial intelligence experts, talk with corporate and startup leaders, explore the ethical and practical implications of AI, and get a sense for how it’s impacting work and life everywhere, as viewed from our perspective on business and technology in Seattle and the Pacific Northwest.
We also plan to continue using and experiencing this new wave of AI ourselves, examining the similarities and differences between neurons and circuits. A great example was this illuminating story by GeekWire managing editor Taylor Soper in March, comparing and contrasting human- and AI-generated poetry.
So it was only natural to apply this experiential and comparative approach to the very process of conceiving the content for this series, using both brains and bots to come up with story ideas.
At our regular news meeting last week, six members of the GeekWire team gathered around a conference room table in Seattle’s Fremont neighborhood. I led the brainstorm from the whiteboard, scribbling as fast as I could (after locating the rare markers that hadn’t dried up after three years of mostly remote work).
Separately, I gave similar prompts and follow-up questions to three generative AI engines: Microsoft’s OpenAI-powered Bing chatbot, Google’s experimental Bard chatbot, and OpenAI’s ChatGPT, powered by GPT-4.
Lessons learned from the process
See below for the detailed output from this experiment in journalistic brains vs. AI bots. But here are some of my high-level takeaways.
- The breadth of story ideas was remarkably similar, spanning a variety of industries and subjects including art, law, music, healthcare, journalism, customer service, ethics, surveillance, and privacy.
- The human brainstorm was more nuanced, often including real-world examples of people and sources we could talk with. It reflected a sophisticated grasp of GeekWire’s journalistic focus, based on many years of experience.
- By comparison, the chatbots took subtle aspects of the prompts too far, seeing things as black-and-white. For example, when prompted to include some ideas for using AI tools, they tended to make every story an experience. I ended up tweaking the prompt language that I used (below) to avoid this.
- The experiment drove home how much more efficient and fast artificial intelligence is than humans in coming up with ideas. For example, ChatGPT took just a few seconds to generate a story list almost as long and detailed as what we created in about a half-hour.
Of course, conducting these processes separately made for a better experiment, to compare AI and humans. But my biggest takeaway, in the end, is that it would have been far more effective to include AI as a participant alongside the human group, querying the chatbots as we went along, to elevate and expand our brainstorm.
Here’s the end result of our whiteboard session.
For more detail, here’s a sample of 10 story ideas from this brainstorm, providing a sense for the breadth and depth of the story ideas we came up with.
Bot-based financial analysis: How well can AI read and summarize an earnings report? This was the source of a flub at Microsoft’s Bing AI chatbot launch, with the AI misreading a Gap Inc. financial statement, but our initial experimentation with various AI tools to analyze breaking earnings reports suggests that there is still some promise here.
Diet, nutrition, and recipes: AI can generate recipes that adhere to specific types of diets and nutritional regimes, but how accurate are they in actually following the principles of these different programs? We would talk with experts in the different diets to find out. And more importantly, how do the resulting meals actually taste?
Fitness and wellness: Along the same lines, how well does AI generate workout routines and fitness regimens adhering to different schools of thought in exercise and wellness? What does the rise of AI mean for personal training?
Bots and music programming: How are radio stations adopting AI as part of their workflows, both nationally and locally? What will AI mean for musical taste and judgment? What do the DJs at Seattle’s KEXP and other independent radio stations think about the rise of artificial intelligence?
Musicians and artificial intelligence: In true brainstorm style, this led our GeekWire team to the natural question of how bands and musicians are thinking about AI as part of their work, building on recent national headlines involving artists such as Drake and Grimes to understand the realities for smaller bands and indie artists.
Can AI make a podcast? We plan to ask different AI engines to script a segment for our weekly GeekWire Podcast, based on specified news stories, and then use AI voices cloned from me and my colleague, GeekWire co-founder John Cook, to read those scripts. We’ve already started working on this, and it has been eye-opening. Stay tuned.
AI and agriculture/horticulture: What happens if we give AI a photo of a tree and ask for advice on pruning? How are farmers in Eastern Washington and elsewhere incorporating the latest in generative AI into their management of fields and operations? Lots of startup angles here, such as AI-powered weed zapping from Carbon Robotics.
Education and AI: Bill Gates, for one, believes that AI will have a big impact on learning, ultimately giving everyone a personalized tutor, customized to individual educational needs. What will this mean for human tutors? And what about the larger issues raised by students using generative AI for assignments? How will the definition of cheating change?
AI in sports: Of course, data analysis is already widely used in sports, but what does the advent of generative AI mean for human judgment in creating day-to-day lineups and preparing players for games? What does it mean for fans and sportsbooks? Our resident sports fans are intrigued and enthusiastic about pursuing these and other angles.
Travel and AI chatbots: This is already a growing phenomenon, raising questions about the future of human travel agents. But how do AI chatbots really stack up against human experts in recommending itineraries and providing guidance to travelers? No surprise, nearly everyone in the room was ready to take on this assignment.
What do the bots say?
Here’s the prompt I gave to the AI chatbots.
The business and technology news site GeekWire is launching a new series called Bot or Not. This special series explores the evolving relationship between humans and machines, and the emerging world of generative and creative AI, examining the ways that robots, artificial intelligence and automation are impacting and fitting into our work and lives. Please brainstorm a list of story ideas, including headlines and descriptions, for GeekWire’s reporters to explore and pursue for this series. Some but not all of the stories should involve reporters using AI tools to understand their capabilities and compare the results to those produced from human creativity, judgment, knowledge and experience. (A recent example was a GeekWire story that compared poetry created by a human with poetry generated by different AI engines from the same prompt.) Also include big-picture stories that explore the implications of AI, not necessarily including hands-on experimentation with AI.
OpenAI ChatGPT: Here is the list generated by OpenAI’s ChatGPT, using GPT-4:
Bot or Not: Can AI Art Rival the Masters? This story investigates the world of AI-generated art and compares it to human-created masterpieces. Reporters will collaborate with AI tools to generate original pieces of art and compare their work to that of renowned human artists, exploring the nuances of creativity, aesthetics, and the ever-evolving relationship between humans and AI.
The AI Lawyer: Weighing the Pros and Cons of Automated Legal Assistance: In this piece, reporters will explore the growing presence of AI in the legal sector. They will use AI-based legal tools to provide advice on a hypothetical case and compare it to the expertise of a human attorney, examining the potential benefits and risks of such technologies in the legal field.
When AI Writes the News: The Future of Journalism in an Automated World: This story will delve into the potential impact of AI-generated news articles on journalism. Reporters will use AI to generate articles on various topics and compare the quality, accuracy, and style to human-written articles, assessing the future role of AI in the newsroom.
Teaching AI Ethics: The Role of Human Values in Machine Learning. In this big-picture piece, the focus will be on the importance of integrating ethics and human values into AI systems. Reporters will explore how developers and researchers are working to ensure that AI technology is developed responsibly and ethically, and the potential consequences if these efforts fall short.
AI in Healthcare: A Deep Dive into the Promises and Pitfalls: This story will explore the use of AI in healthcare, from diagnostics to treatment planning. Reporters will test AI-driven diagnostic tools and compare their accuracy and effectiveness to human medical professionals, examining the potential benefits and drawbacks of incorporating AI in patient care.
AI-powered Customer Service: Efficiency vs. Empathy: Description: In this piece, reporters will interact with various AI-powered customer service tools, comparing their efficiency and accuracy to human customer service representatives. The story will also address the balance between automation and empathy, and whether AI can truly replace human interaction in this area.
The AI Music Revolution: Composer or Collaborator? This story will investigate the role of AI in the world of music composition and production. Reporters will collaborate with AI tools to create original music and compare the results to human compositions, exploring the potential for AI to become a valuable creative partner in the music industry.
Safeguarding Privacy in the Age of AI Surveillance: This big-picture piece will examine the implications of AI-driven surveillance technologies on individual privacy and civil liberties. Reporters will delve into the current state of AI surveillance, its potential impact on society, and the measures needed to protect privacy in an increasingly monitored world.
AI and the Future of Work: Job Creation or Destruction? Description: In this story, reporters will explore the implications of AI and automation on the job market. They will examine the potential for AI to create new job opportunities while also displacing existing roles, and discuss the need for reskilling and workforce adaptation in the face of rapid technological change.
The AI Dating Game: Can Machines Help Us Find True Love? This piece will delve into the world of AI-driven dating apps and services. Reporters will use AI tools to help them find potential matches and compare the results to traditional human-based matchmaking, exploring the role of AI in modern dating and relationships.
Google Bard: This was the best (in my opinion) of three drafts from Google’s Bard.
Can AI Write Code? As AI technology continues to develop, it is becoming increasingly capable of performing tasks that were once thought to be the exclusive domain of humans. One such task is writing code. In recent years, there have been a number of AI-powered code generators that have been developed, and some of these tools have already been used to create successful software products. In this story, GeekWire reporters could explore the capabilities of AI code generators, and they could also interview developers who have used these tools to create their own software.
AI-Powered Chatbots Are Changing the Way We Communicate: AI-powered chatbots are becoming increasingly common, and they are already having a significant impact on the way we communicate. Chatbots are being used in a variety of settings, including customer service, sales, and education. In this story, GeekWire reporters could explore the different ways that chatbots are being used, and they could also interview people who have interacted with chatbots.
AI Is Changing the Future of Work: AI is already having a significant impact on the workforce, and this impact is only going to grow in the years to come. As AI-powered machines become more capable, they will be able to automate more and more tasks that are currently performed by humans. This could lead to significant job losses in some industries, but it could also create new jobs in other industries. In this story, GeekWire reporters could explore the potential impact of AI on the workforce, and they could also interview people who are already being affected by AI-driven automation.
The Ethics of AI: As AI technology continues to develop, it is important to consider the ethical implications of this technology. Some experts have raised concerns about the potential for AI to be used for malicious purposes, such as creating autonomous weapons systems. In this story, GeekWire reporters could explore the ethical implications of AI, and they could also interview experts who have thoughts on this topic.
Microsoft Bing: And here is the list from the Bing AI chatbot:
“How AI is changing the way we work and live” – This story could explore how AI is changing the way we work and live, including how it is impacting our jobs, our homes, and our daily lives.
“The ethics of AI” – This story could explore the ethical implications of AI, including how it is being used in different industries and how it is affecting our society as a whole.
“The future of AI” – This story could explore what the future holds for AI, including how it will continue to evolve and what new applications we can expect to see in the coming years.
“AI and creativity” – This story could explore how AI is being used to create art, music, and other forms of creative content.
“AI and healthcare” – This story could explore how AI is being used in healthcare, including how it is being used to diagnose diseases and develop new treatments.
“AI and education” – This story could explore how AI is being used in education, including how it is being used to personalize learning and improve student outcomes.
“AI and the environment” – This story could explore how AI is being used to address environmental challenges, including climate change and pollution.
“AI and cybersecurity” – This story could explore how AI is being used to improve cybersecurity, including how it is being used to detect and prevent cyber attacks.
“AI and transportation” – This story could explore how AI is being used in transportation, including how it is being used to improve traffic flow and reduce accidents.
“AI and customer service” – This story could explore how AI is being used in customer service, including how it is being used to improve customer experiences and reduce wait times.
So many ideas! Now we just need to get these bots to report and write the stories.
Just kidding … or maybe not.
Which list do you prefer? What would you add? Email tips@geekwire.com with your thoughts and feedback. Thanks for reading, and look for more “Bot or Not” soon.
