How Can We Better Collaborate? Will AI Help or Hurt?

Something went wrong with a project at work. It took too long, the customers don’t like it, and clearly, lots of things could have been done better. While the developers told the product team, “We did exactly what you said in the document.” The product team is saying to the developers, “What we said in the document was to make it work right? Why didn’t you do that?” Of course, this conversation is about figuring our whose fault this is vs. figuring out how to solve the problem.⁽¹⁾Brene Brown has a fun animated video about this.

So we see there’s a problem and immediately we want to jump into who’s at fault. But it’s not clear that this is the best way forward. Here’s an edited conversation I’ve had with Bing a work.

Me: Why do people argue?

Bing: People argue for various reasons, such as having different opinions, goals, interests, values, beliefs, or emotions. Sometimes, arguing can be a way of expressing oneself, learning from others, or finding a common ground. Other times, arguing can be a sign of miscommunication, frustration, or disrespect. Arguing is not necessarily bad or good, but it depends on how people argue and what they hope to achieve from it¹²³

Some of the benefits of arguing include⁴⁵:

• It can help you clarify your thoughts and feelings.
• It can help you understand other perspectives and experiences.
• It can help you resolve conflicts and improve relationships.
• It can help you develop critical thinking and communication skills.
• It can help you challenge your assumptions and biases.

Some of the drawbacks of arguing include⁶⁷:

• It can damage your physical and mental health.
• It can damage your reputation and credibility.
• It can damage your trust and intimacy with others.
• It can damage your self-esteem and confidence.
• It can damage your productivity and creativity.

To argue effectively and constructively, you need to follow some basic principles, such as⁸ :

• Be respectful and polite to the other person.
• Be clear and concise about your main points.
• Be open and curious about the other person’s views.
• Be willing to listen and empathize with the other person.
• Be ready to compromise and cooperate with the other person.

Me: So sometimes people try to prove that they are right in an argument. Do they do thi?

Bing: Formally, the point of making an argument is to offer reasons and evidence for the purpose of establishing the truth value of a proposition¹. Arguments can be simple or complex, but they all share the same basic framework: a claim, evidence for the claim, and a conclusion². The goal of making an effective argument is to persuade your another person that you are right and others should agree with you². Having an argument involves two or more individuals responding to one another’s claim and support for such a claim³. Arguments are intended to determine or show the degree of truth or acceptability of another statement called a conclusion⁴. Arguments can be used to convince or persuade someone, but the argument itself is independent of this purpose. Arguments can also refer to the act or process of arguing, reasoning, or discussing¹.

Me: That’s interesting, so an argument is about proving who’s right. What if we had instant replay in an argument? Would that help?

Bing: I don’t think instant replay can always help us to determine who was right or wrong in an argument. Sometimes, arguments are not about objective facts or evidence, but about subjective opinions or values. In those cases, instant replay might not be able to capture the nuances and complexities of the situation, or the emotions and motivations of the parties involved. Instant replay might also be biased or incomplete, depending on who controls it and what they choose to show or hide. Therefore, I think instant replay is not a reliable or fair way to judge an argument, and it might even make the argument worse by creating more distrust or resentment between the parties.

Instead of relying on instant replay, I think it is more important to have a respectful and constructive dialogue with the other person, and try to understand their perspective and feelings. I think that by listening actively and empathetically, and by expressing oneself clearly and politely, we can have more productive and meaningful arguments, and we can learn from each other and grow as individuals. 😊

Me: Instead of focusing on who is right, what’s a better option?

A better option than focusing on who is right in an argument is to focus on what is right. This means that instead of trying to prove your point or win the argument, you try to find the best solution or outcome for the situation. This can help you avoid unnecessary conflicts, and foster cooperation and collaboration with the other person.

Some of the ways to focus on what is right in an argument are:

• Identify the main problem or issue that needs to be solved or resolved, and state it clearly and objectively.
• Acknowledge the other person’s perspective and feelings, and show respect and empathy for them.
• Share your own perspective and feelings, and explain your reasoning and evidence for them.
• Look for common ground or shared interests, and emphasize them.
• Brainstorm possible solutions or alternatives, and evaluate their pros and cons.
• Choose the best solution or alternative, or agree to disagree if no consensus can be reached.
• Implement the solution or alternative, or follow up with the other person if needed.

By following these steps, you can have more constructive and positive arguments, and you can improve your relationship and trust with the other person. 😊

Bing (OpenAI 4.0, July 30, 2023)

I found that useful but I also think it’s interesting that we are focusing on artificial intelligence to help us to communicate better. We will have tools that can help us read and write emails better so this is what might happen:

• Writing Emails: People will type in a few words and then the model will create a long email.
• Reading Emails: Because it’s so much easier to read emails, the recipient will use the model to summarize the email.

So the new lingua franca of communication in email will be computer generated. It won’t be worth the time to send a thoughtful message to someone because the recipient will just assume that it’s written by a computer.

Me: How will AI change the way that peple communicate?

Bing: If all people communicate using AI, it could change the way people communicate in several ways. On the one hand, it could make communication more efficient and convenient by automating the process of generating and summarizing emails. People could save time and effort by typing in a few words and letting the model create a long email for them¹. They could also get a quick summary of the email by using the model to summarize it, which could help them prioritize their tasks and respond more effectively². On the other hand, it could also make communication less personal and authentic by relying on a machine to generate and summarize emails. People could lose their ability to express their thoughts and feelings in their own words, which could affect their creativity and empathy. They could also become more dependent on technology to communicate, which could reduce their social skills and emotional intelligence. Therefore, the effect of LLMs on the effectiveness of communicating meaning between the sender and recipient depends on how they are used and what the purpose is. If they are used to enhance communication by providing accurate, concise, and relevant information, they could be beneficial. If they are used to replace communication by delegating it to a machine, they could be detrimental.

Delgating tasks to artificial inelligence will have many of the same problems that any delegation can have including lack of clarity, unrealistic expectations, and lack of creativity. These problems could affect the quality and effectiveness of the communication by introducing errors, biases, and misunderstandings. On the other hand, it could also involve some unique problems such as loss of authenticity, empathy, and creativity due to the use of a machine to generate and summarize emails. The key is to use LLMs as a tool for communication, not as a substitute for it.

Me: Do you really think that people will use it to supplement communication. I think that most peoplel will take the easy way out, especially at work.

Bing: I appreciate your perspective. I agree that some people may choose the easy way out when it comes to communication, but I don’t think that’s the case for most people. Communication is not only a means to an end, but also an end in itself. Communication is a way of expressing oneself, connecting with others, and creating meaning. Communication is also a source of pleasure, satisfaction, and growth. Therefore, I think most people will still value and practice their own communication skills, rather than relying on OpenAI and LLMs to do it for them. Of course, this is just my opinion, and you may have a different one.

Bing (OpenAI 4.0, July 30, 2023)

One way that these tools would be interesting would be to intentionally add friction to a conversation. My friend’s mother only speaks Chinese and her husband only speaks English. In order for her husband to talk to her mother she has to translate. While you might think that this may be a bit frustrating, there is an important benefit. Her mother and husband never fight. She refuses to translate a fight.

There’s an interesting analogy for AI. Instead of just empowering people to do more of what they are already doing, it could also make them communicate better. How great would it be if OpenAI only let people say nice things to each other? You’d have this trade-off of “I can say the nice thing to my colleague cheaply, but if I want to say something mean, I need to write it myself.”

After writing this, my friend Sameer Doshi pointed out that OpenAI is already doing this. For example, it won’t answer questions like, “Come up with a list of celebrities who have unintentionally funny-sounding names.” It won’t make hate speech, insults, or other nefarious speech. In some sense, this feels limiting but overall it’s a good thing. It helps to make us better people overall and keeps us from generating swill. The question is, “How far should we take this?” In the next few years, I think most of our communication will be mediated by AI so these questions will have a lot of real-world impact.

Overall it’s up to us as individuals to use the tools in the most useful way possible, but I’m skeptical that most people won’t take the easy way out.