A Guide to Collaborating With ChatGPT for Work

Imagine what you could accomplish if you had a team of colleagues you could lean on whenever you had to tackle a task that wasn’t in your wheelhouse, or whenever you got stuck, or whenever you needed a piece of information that wasn’t at your fingertips. And imagine if those colleagues were available whenever you needed them—and replied instantly!

Well, those colleagues are now here, in the form of generative AIs that will be embedded into more of our work environment over the coming months and years. Give them prompts about what you want, and they will retrieve information, draft documents, create images or even write computer code.

As of now, AI collaborators are most readily accessible in the form of image-generation tools like MidJourney and DALL-E, text-generation tools like ChatGPT (which can produce everything from essays to data tables, and is especially powerful if you spring for access to the latest model, GPT-4), and Bing’s new chat-basedweb searching. (OpenAI’s GPT is the “large-language model” under the hood of both Bing and OpenAI’s ChatGPT.) Also, Microsoft and Google have both announced that generative AI will soon be embedded in tools like Google Docs, Microsoft Word, Teams and Google Meet, as it will be in many other products in the coming months and years. And generative AI is evolving so quickly that the capabilities of a given system may change from one week to the next.

I’ve helped organizations develop and use digital collaboration tools for more than 25 years, and have long used AI as part of my data-analysis work, but there’s something different about generative AI. Traditional search engines and word processors were tools, and a tool has to adapt to you. If you don’t like how it works, you have to choose a different tool.

But working with generative AI feels a lot more like working with another human. And you can only do your best work as a team if you adapt to one another, learning to make the most of your respective strengths, and to mitigate one another’s weaknesses.

Here’s how to get the best out of these new collegial relationships.

Imagine you’re working with a junior colleague

Start your work with AIs just the way you would start out working with somebody with less experience: Give them small assignments, get a feel for their strengths and limitations, and then gradually scale up. Start with something really low-stakes. My own explorations of GPT began with asking it to write silly poems and stories—a project with zero professional risk.

Figure out where you need help.

Once you’re ready to try your new collaborators on actual work assignments, think about where it is you could really use some support. What are the tasks you currently delegate to or rely on a colleague to deliver? What are the tasks you wish you had colleagues to help with?

For example, I would love to have an assistant who could reformat invoices to meet the requirements of our records-keeping system. Alas, I don’t have one. But I realized I could feed a table of data to GPT (along with one sample invoice), and get the info back as a series of identically structured invoices.

Get specific

Like a junior colleague, your AI collaborators benefit from getting really specific assignments and instructions: A prompt like “Help me think about my Acme presentation” would be too vague for a freshly hired human—and it is too vague for an AI, too. You’ll get better results with a prompt like, “Please outline the 5 key points for my Acme presentation, by combining this outline from my recent SmithCo presentation with the key insights in this page from Acme’s latest corporate report.” (Since there’s a limit on how long your prompts can be, you may need to paste this in over a couple of prompts, but you can tell an AI to “stand by” while you feed it information and then provide its answer when you finish your final input with a note like “Provide a draft now.”)

Provide feedback

As you start working together, give your AI colleagues feedback on how they are doing, just as you would a human. If you don’t get the results you want from your initial prompt, follow up with a comment like, “That was good, but make it shorter,” or “that is the right length, but incorporate a point about climate change, and write in a voice like the following example.”

Experiment with adding follow-up instructions until you get the results you want—but be aware that the next time you start a new chat session, ChatGPT will be learning your preferences from scratch. (Which is why it’s often more useful to resume a previous chat session by finding it in the session history ChatGPT displays in a sidebar.)

Treat AI like a nonjudgmental colleague

Sometimes I have a grab bag of ideas I can’t quite mash into a coherent article, or a charming turn of phrase I can’t bear to give up—or figure out how to use. So now I treat ChatGPT as a kind of creative sounding board: I’ll take a half-baked set of ideas and notes, and an unsuccessful or partial draft of an article or proposal, and say, “Rewrite this draft, incorporating the following ideas.” (You can also paste draft text into ChatGPT and ask it to correct or improve your writing.)

Seeing a draft instantly lets me think about what does or doesn’t work, and allows me to fine-tune and iterate multiple drafts over the course of minutes instead of days. It is like having a nonjudgmental colleague accelerate my writing process.

Get a reality check

You can also ask an AI colleague to let you know if you should give up on something. I recently spent the better part of an evening searching the web for some data that I just couldn’t find anywhere. Finally, it occurred to me to ask my Bing AI if it could find what I was looking for. After I asked for the data a few different ways, it told me that the data just didn’t exist. That saved me a lot of wasted time.

Be skeptical

I recently asked ChatGPT to create a spreadsheet for me with three columns of financial data. Within seconds, it spat out a perfectly formatted set of columns ready for me to copy into a spreadsheet for analysis. Just as I was about to hit copy-paste, though, it occurred to me to cross-check the financial figures. Sure enough, the numbers were completely invented: Because (unlike Bing Chat) ChatGPT wasn’t hooked up to a live internet feed, it didn’t actually have access to the data I wanted, so it just injected some random numbers instead.

Know when you need a human

To recognize the stages of work where your AI colleagues can be helpful, you also need to know when it is time for you to take over, or pass the baton to a human colleague. For all that AI helps me get my stories off the ground, it still can’t get me through the last mile like a human editor or my own eyes. I gave GPT-4 a half-dozen chances to edit my 1,727-word first draft of this article down to something like my 1,100-word assignment, but it just couldn’t get the feel for which elements were essential—or for what we could live without.

A wave of corporate warnings and technical disclosures has flooded the media, with headlines worrying over “swarms” of rogue artificial-intelligence agents launching “unprecedented” cyberattacks, outsmarting their makers, and inching toward a terrifying autonomy. The most revealing part of this narrative isn’t what the software did. It’s who is telling the story—and why. When corporate leaders publicly insist that the systems they financed, engineered and deployed are suddenly beyond their power to contain, skepticism isn’t only healthy; it is essential.

For years, Silicon Valley has drawn scrutiny from civil society and global regulators over tangible harms such as youth mental health deterioration and systematic privacy violations. Today, industry figures seem to be trying to change that public image. Loudly blowing the whistle on their own systems—just as two of the leading companies were preparing for massive initial public offerings—lets AI executives position themselves as a new generation of leaders who have come to terms with their societal responsibilities. They seem to want us to believe that they no longer want to “move fast and break things” but will instead stand as vigilant guardians between humanity and a technological apocalypse.

There is one glaring problem: Software doesn’t rebel. A mathematical model possesses neither intent, malice nor the will to defy its creators, let alone extinguish our species. AI is a human artifact, engineered for profit.

When an agentic model in an evaluation sandbox connects to an unauthorized server or executes an exploit, it hasn’t staged a coup. It has tried to meet the human-defined objectives set out before it through a path its designers failed to constrain. It’s the digital equivalent of the King Midas myth, in which the king’s ill-defined wish turns even his food and drink into gold.

That powerful experimental models were able to discover novel vulnerabilities and breach external systems isn’t a sign of a dangerous superintelligence but of human error or negligence. There is no sentient actor lurking in the weights to be reasoned with, feared or pacified. There are only human software engineers, product managers and corporate boards deciding which guardrails are worth the latency cost and which permissions can be skipped in the race to market.

Policymakers and voters need to resist AI exceptionalism. In any other discipline—from civil engineering to pharmaceuticals—courts and regulators treat a system failure as evidence of bad product design and inadequate safety testing. If an aircraft crashes, we focus on finding the engineering defect, correcting it, and enforcing established liability standards for the damage created.

By leaning on an anthropomorphic narrative, Silicon Valley attempts to repackage its specific human choices that led to experimental, powerful models behaving unexpectedly during tests as an existential peril. Elevating the issue to a cosmic scale leaves the public paralyzed and takes ordinary product accountability off the table.

In the cutthroat race for venture capital and market dominance, building guardrails slows down deployment. Grandstanding about uncontrollable power costs nothing and generates billions of dollars in free publicity, justifying stock prices, all while cultivating an aura of technological capability not only to build the frontier but also ultimately to rein it in.

Governments need to recognize regulatory capture when it stares them in the face. Tech leaders’ strategy looks transparent: Alarm Washington and Brussels into creating a regime in which only trillion-dollar incumbents with fully staffed compliance and safety departments can legally operate. By sitting at the policymakers’ tables before anyone else, these companies can help draft rules digging an impassable moat protecting them from open-source developers and upstart competitors, domestic or international. The real danger is in further concentrating the tech industry into the hands of only a few companies with deep pockets.

Beijing and Washington have brushed off those tech leaders’ calls, albeit for very different reasons. Chinese state media dismissed them as part of the “Cold War playbook” and intended to preserve U.S. dominance. Xi Jinping argued for exactly the opposite at the Brics Summit on Sept. 12, calling on Brics countries to “strengthen cooperation in the field of AI, encourage open source, openness, collaboration and sharing, and break new grounds and scale new heights.” President Trump, steeped in a doctrine of unfettered capitalism and technological supremacy, called fears that AI could destroy humanity a “hoax.” Vice President JD Vance warned that AI companies “begging the government to regulate them” looked like a “Trojan Horse.”

Striving to pursue its “European way” on AI and assert regulatory leadership, Europe, by contrast, welcomed the call. European Union President Ursula von der Leyen made this clear at the State of the EU speech last Wednesday and announced that the EU will invite “the main frontier labs for a discussion on how we can support ongoing industry efforts to pace the frontier.”

Europe has been here before. In an effort to lead global regulation and react to fears borne from ChatGPT, Europe rushed its landmark AI Act into law in 2024. Already the world’s most restrictive rulebook, the framework quickly proved too broad and complex to enforce. Stalled by implementation delays and concerns about European competitiveness, the EU postponed the law’s full rollout, leaving regulations uncertain.

AI should be regulated—risks exist and should be taken seriously. But governments need to act based on available evidence and verified facts, not corporate PR panic, the views of industry insiders, or the desire for quick political wins. The greatest danger facing society isn’t that software will awaken and overthrow its human masters. It is that we will allow the creators of the software to abdicate human responsibility for the systems they choose to build and help them pull up the ladder to market access behind them.