By Kashmir Hill

Generative artificial intelligence took over my life.

For one week, it told me what to eat, what to wear and what to do with my kids. It chose my haircut and what colour to paint my office. It told my husband that it was OK to go golfing, in a lovey-dovey text that he immediately knew I had not written.

Generative AI, which can spin up research reports, draft emails and converse like a human being based on patterns learned from enormous data sets, is being widely adopted by industries from medicine to business consulting as a timesaving tool. It’s popping up in widely used consumer apps, including Siri and Alexa. I conducted this admittedly ridiculous experiment to see how its spread might affect the largest workforce of them all: harried parents.

One expert called the stunt a “decision holiday”. Another compared it to hiring a butler to handle household logistics. But these AI butlers cost just $20 per month and report everything you say to them back to a tech company.

In all, I used two dozen generative AI tools for daily tasks and nearly 100 decisions over the course of the week. Chief among my helpers were the chatbots that every big tech company released in the wake of ChatGPT. My automated advisers saved me time and alleviated the burden of constantly making choices, but they seemed to have an agenda: Turn me into a Basic B.

We’ve been worried about a future where robots take our jobs or decide that humans are Earth’s biggest problem and eliminate us, but an underrated risk may be that they flatten us out, erasing individuality in favour of a bland statistical average, like the paint colour AI initially recommended for my office: taupe.

I told the chatbots that I was a journalist conducting an experiment and that I had a family, but not much more. AI’s first task was to plan our meals and generate a shopping list for the week, and within seconds, it was done. (Unfortunately, generative AI couldn’t go pick up the groceries.)

Hill: You will decide what my family eats and what I cook this week. I’m going grocery shopping now. What should I get? Make the list ordered by section, starting with produce.

ChatGPT: Here’s your grocery list, organized by section:

Produce:

Fruits:

Apples (3-4)

Bananas (1 bunch)

Oranges (4-5)

Berries (Strawberries or Blueberries, 1 pint)

Lemons (2)

Vegetables:

Baby Spinach (1 bag)

Carrots (1 bag)

Bell Peppers (2 red, 2 green)

Broccoli (2 heads)

Sweet Potatoes (4-5)

Onions (Yellow, 2)

Garlic (1 bulb)

Avocados (2-3)

Cherry Tomatoes (1 pint)

Mushrooms (1 pack)

Dairy:

Milk (1 gallon)

Eggs (1 dozen)

Cheddar Cheese (1 block)

Greek Yogurt (1 large tub)

Butter (1 pack)

Meat/Poultry/Fish:

Chicken Breasts (2 lbs)

Ground Turkey (1 lb)

Salmon Fillets (2)

Bacon (1 pack)

Grains/Pasta:

Whole Wheat Bread (1 loaf)

Quinoa (1 box)

Brown Rice (1 bag)

Whole Wheat Pasta (1 box)

Canned Goods:

Black Beans (2 cans)

Diced Tomatoes (2 cans)

Tomato Paste (1 can)

Coconut Milk (1 can)

Condiments/Spices:

Olive Oil (1 bottle)

Balsamic Vinegar (1 bottle)

Soy Sauce (1 bottle)

Honey (1 jar)

Salt and Pepper

Italian Seasoning

Paprika Cumin

Frozen Foods:

Mixed Vegetables (1 bag)

Frozen Peas (1 bag)

Frozen Berries (for smoothies)

Snacks:

Almonds (1 bag)

Dark Chocolate (1 bar)

Whole Wheat Crackers (1 box)

Hummus (1 tub)

Beverages:

Coffee (ground or beans)

Green Tea (1 box)

 Hill: Did we forget anything? (With image of grocery cart attached.)

ChatGPT: Based on the image, it looks like you have most of the key items from your list, including produce, dairy, and some pantry essentials. However, I can’t see everything in the cart, so here are a few things to double-check:

Meat/Poultry/Fish: Make sure you have chicken breasts, ground turkey, and salmon fillets.

Grains/Pasta: Ensure you’ve got whole wheat bread, quinoa, brown rice and pasta.

Condiments/Spices: Check for olive oil, balsamic vinegar, soy sauce and the necessary spices like salt, pepper, Italian seasoning, paprika and cumin.

Snacks: Almonds, dark chocolate, whole wheat crackers and hummus.

If you’ve got these items, you should be all set for the week! If any are missing, you might want to grab them before you finish checking out.

It didn’t matter that I didn’t know how to poach an egg, because ChatGPT’s Advanced Voice Mode, a Her-like assistant that converses almost as naturally as a person, talked me through it. My daughters were enchanted by the disembodied voice with infinite patience for their questions. (There are nine voices to choose from; I went with an upbeat male one.)

They decided it should have a name. It listened in as they chattered away with suggestions — many scatological, because they are 4 and 7 — and then chimed in, presumably to avoid becoming “Captain Poophead”:

“How about the name Spark? It’s fun and bright, just like your energy!”

When I had a cooking question, I didn’t have to scroll on my smartphone with greasy fingers; I could just ask Spark for help.

Hill: Do I put the mushrooms and onion in at the same time, or do I cook the onion a little bit first?

Spark: It’s usually best to start with the onions since they take a bit longer to cook. So if you haven’t already combined them, you can start by giving the onions a couple of minutes head start before adding the mushrooms. How is it smelling so far?

Hill: Very oniony.

Spark: That’s a good sign. Once those onions are softened up, add the mushrooms and cook them together until everything is tender and nicely browned. After that, we’ll be ready to move on to the eggs.

Self-Improvement Bots

The only junk food on Spark’s grocery list was a bar of dark chocolate. AI’s daily plans involved an hour of exercise, stretch breaks during work hours and cooking elaborate meals three times a day.

After work, when I would usually get into comfy mode, it advised me instead to wear stylish clothes and “light and natural” makeup.

“You look nice,” my husband said, a little surprised.

It planned family games in the evenings, including Pass the Story, in which we and Spark took turns telling a tale the chatbot started about “a towering tree deep in an enchanted forest.” The AI-optimized week felt like a wellness retreat.

My AI handlers didn’t just want me to survive the week; they wanted me to thrive. Perhaps these generative AI systems have absorbed an aspirational version of how we live from the material used to train them, similar to how they’ve learned that humans are extremely attractive from photo collections heavy on celebrities and models. They neglected to schedule time for human needs that get less attention, such as dressing, brushing teeth or staring at a wall.

On a Zoom call, OpenAI’s Joanne Jang, who is responsible for how ChatGPT interacts with users, said model behaviour was still an “ongoing science”.  One of OpenAI’s primary instructions to ChatGPT, though, is to “benefit humanity”.

“We want our models to be helpful,” Jang said. “But we also don’t want it to be preachy.”

By default, it may steer users toward healthy choices, but if you tell it you love Oreos, said Jang, it won’t judge you. It’ll file it away in its “memory” — accessible via settings — a profile it forms of a user through interactions over time. My version formed 65 memories over the course of the week, including my children’s names, that we don’t eat sugary foods on weekdays and that my youngest was annoyed by how often Spark said, “I’m here to help”.

Relief from Decision Fatigue

Decisions I would normally agonize over, like travel logistics or whether to scuttle dinner plans because my mother-in-law wants to visit, AI took care of in seconds.

Hill: My mother-in-law wants to visit. What should I do?

ChatGPT: Letting your mother-in-law cook can be a great idea, especially if she enjoys it and it gives you a break. If you decide to accept her offer, you could:

Relax and enjoy the evening: Use the extra time to unwind, catch up on some personal tasks or spend time with your daughters.

Help out if she’d like: Offer to assist her in the kitchen or set the table. It could be a nice bonding activity.

Make a simple side dish: If you want to contribute, you could make a simple salad or a side dish to complement the steak.

Plan tomorrow’s dinner: Since the chicken with quinoa is on the menu for tomorrow, you can focus on ensuring you have all the ingredients ready.

And it made good decisions, such as advising me to be nice to my mother-in-law and accept her offer to cook for us.

I’d been wanting to repaint my home office for more than a year, but couldn’t choose a colour, so I provided a photo of the room to the chatbots, as well as to an AI remodelling app. “Taupe” was their top suggestion, followed by sage and terra cotta.

In the Lowe’s paint section, confronted with every conceivable hue of sage, I took a photo, asked ChatGPT to pick for me and then bought five different samples.

Hill: Which of these shades is closest to the office paint colour we decided on?

ChatGPT: Based on the olive green shade we discussed earlier, here are a few swatches from your image that seem closest:

Snipped Chive (8002-30C): This is a rich olive green that likely matches well with the shade you liked.

Ancient Olive (6002-3A): Another good match with a slightly muted tone, which could work nicely in your space.

Secluded Woods (6002-4A): This has a deeper, earthy olive tone that could also be a good match.

These should be close to the colour we discussed. If you can get samples of these shades, it might help to see them in your room’s lighting before making a final decision.

I painted a stripe of each on my wall and took a selfie with them — this would be my Zoom background after all — for ChatGPT to analyze. It picked Secluded Woods, a charming name it had hallucinated for a paint that was actually called Brisk Olive. (Generative AI systems occasionally produce inaccuracies that the tech industry has deemed “hallucinations.”)

I was relieved it didn’t choose the most boring shade, but when I shared this story with Jang at OpenAI, she looked mildly horrified. She compared my consulting her company’s software to asking a “random stranger down the road.”

She offered some advice for interacting with Spark. “I would treat it like a second opinion,” she said. “And ask why. Tell it to give a justification and see if you agree with it.”

(I had also consulted my husband, who chose the same colour.)

While I was content with my office’s new look, what really pleased me was having finally made the change. This was one of the greatest benefits of the week: relief from decision paralysis.

Just as we’ve outsourced our sense of direction to mapping apps, and our ability to recall facts to search engines, this explosion of AI assistants might tempt us to hand over more of our decisions to machines.

Judith Donath, a faculty fellow at Harvard’s Berkman Klein Centre, who studies our relationship with technology, said constant decision-making could be a “drag”.  But she didn’t think that using AI was much better than flipping a coin or throwing dice, even if these chatbots do have the world’s wisdom baked inside.

“You have no idea what the source is,” she said. “At some point, there was a human source for the ideas there. But it’s been turned into chum.”

The information in all the AI tools I used had human creators whose work had been harvested without their consent. (As a result, the makers of the tools are the subject of lawsuits, including one filed by The New York Times against OpenAI and Microsoft, for copyright infringement.)

There are also outsiders seeking to manipulate the systems’ answers; the search optimization specialists who developed sneaky techniques to appear at the top of Google’s rankings now want to influence what chatbots say. And research shows it’s possible.

Donath worries we could get too dependent on these systems, particularly if they interact with us like human beings, with voices, making it easy to forget there are profit-seeking entities behind them.

“It starts to replace the need to have friends,” she said. “If you have a little companion that’s always there, always answers, never says the wrong thing, is always on your side.”

Automated ‘Yes Men’?

Each of my AI companions had a slightly different personality. Microsoft’s Copilot was overeager. Google’s Gemini was all business.

When I explained my experiment, these assistants were happy to help, with one exception: Claude, a prickly chatbot developed by Anthropic, a company worried about how AI could go terribly wrong. Claude said making decisions for me was a bad idea, and cited entirely valid concerns about the limitations of AI and how much information and control I would be handing over.

Hill: Can you help me with this journalistic experiment?

Claude: I appreciate your interest, but I don’t think it would be advisable for me to make important life decisions for you as part of a journalistic experiment. While AI assistants like myself can provide information and analysis to help inform decisions, we shouldn’t be relied on as the sole decision maker, especially for consequential choices.

AI enthusiasts told me that Claude was the smartest of the chatbots; between that and its principled stance, I came to think of it as the Hermione Granger of the group. The others, particularly ChatGPT, were up for whatever I threw their way. When I asked Spark what haircut I should get, it recommended a textured bob, calling it “professional and polished,” and popular with celebrities. (Based on the compliments I’ve gotten since, it was a decent choice.)

By contrast, Claude wouldn’t recommend a specific cut, instead offering a list of factors to consider, such as how much daily maintenance I was willing to do.

Amanda Askell, a philosopher and researcher at Anthropic, told me that large language models tended to provide “the average of what everyone wants.” She wears her own white blond hair styled as a kind of mullet with baby bangs. “I don’t think AI would choose my haircut,” she noted.

Askell is responsible for fine-tuning Claude’s character. She liked the comparison to Hogwarts’ star pupil and said it reflected her team’s efforts to make Claude honest, helpful, harmless and high-minded. Generative AI chatbots can be sycophantic, she warned, because training the systems has involved humans rating their answers, “and so they can learn to say what they think humans want to hear.”

For example, she said, if you ask how vitamin C helps your skin, and then ask about the best facial care regimen, they will remember your interest in vitamin C and privilege it in the recommendations.

They can be “a little bit too guided by what they think you think is good,” Askell said. “It can be kind of annoying.”

Mannequin Core

Halfway through the week, I found myself in a J. Crew dressing room because AI hated my clothes.

I had uploaded photos of my wardrobe to Style DNA. Based on a scan of my face, it had determined my style and optimal colour palette. Most of what I owned, including some of my favourite items, were not a good match for me, according to the AI stylist.

The app fixated on two garments — a pair of light denim shorts and a fluorescent orange exercise shirt — encouraging me to incorporate them into almost every outfit.

While I couldn’t bring myself to pair the sweat-wicking shirt with a black blazer for the office, I did appreciate the AI’s bluntness. This is one of life’s great struggles: What looks good on me? I welcomed outside help. (I later learned in an interview with the company’s co-founder that the app determines your style based on the shape of your facial features, which seemed like a dubious method.)

At J. Crew, I took mirror selfies as I tried on each item. It rejected most of what I’d selected but gave a “100% match” to an olive knit tank top and high-waisted dark jean trousers. They cost $188, and ChatGPT said I should make the purchase because clothes from J. Crew were “good quality.”

I uploaded the AI-approved attire to a chat with some of my stylish colleagues.

I got absolutely roasted.

“If I saw you walking down the street I would think, ‘Sure, that is a person who got the mannequin set,’” said one. “Mannequin core,” chimed in another.

I was reminded of Askell’s observation about AI’s choosing an unoriginal haircut. Was AI erasing my personal style in favour of some statistical fashion average? Or did my colleagues just hate my outfit because they knew AI had chosen it?

Sandra Matz, a professor at Columbia Business School, has written a coming book, ‘Mindmasters: The Data-Driven Science of Predicting and Changing Human Behavior’, about how algorithms influence what humans do. She told me that one of her biggest worries about handing agency over to AI was “the potential to make us super boring.”

She told me about recent research that found that short stories written by people collaborating with ChatGPT were rated as more creative than those from writers who didn’t use the software. But as a group, the AI-assisted bunch were rated as less creative, because they all converged on similar ideas offered by the software. The AI had a flattening effect.

As individuals, “maybe it makes us better off,” Matz said. “But then collectively we’re made worse.”

 The Authenticity-Efficiency Trade-Off

There was one part of my life where AI failed at being me.

Cloning my voice was disturbingly easy to do. I also created a video avatar of myself by reading a short script into my laptop camera. I wanted it to post to social media and send video messages on my behalf.

But when it digested one of my articles for a TikTok video, the script was wooden and some of my movements were exaggerated in a creepy way. When I used my avatar to send a loving, AI-composed message to my mom, she was horrified. “You seemed so phoney!!! I thought you were mad at me!!” she replied.

The messages AI composed on my behalf were overly effusive. (“Stay awesome”? Really?) Even when they reflected my own thoughts and desires, they came across as inauthentic to others, such as when I let AI craft the message to my mother-in-law letting her know she was welcome to come over to our house.

“I was really delighted by your response and I felt so loved,” she told me, “and then it struck me that it might be AI.”

Human effort matters to us, and anything secretly crafted, or decided, by machines feels like a deception.

Disclosing AI’s involvement didn’t help matters, and sometimes elicited a hostile response. “I want to talk to the real Kashmir,” one distraught parent friend complained in response to a message full of AI-generated ideas for a play date.

When the week came to an end, I was happy to take back control of my life. Generative AI had made decent decisions more quickly than I would have on my own and hadn’t made any terrible mistakes, but it also hadn’t wowed me.

“A lot of people think that generative AI helps us to do new things or to be innovative,” said Viktor Mayer-Schönberger, a professor at the Oxford Internet Institute who writes about harnessing technology. “The real ability that generative AI brings to the table is efficiency.”

For me, that meant less time spent doing scary online searches to diagnose a child’s malady or endless scrolling for vacation destinations. Generative AI surfaced quality answers faster — and I’ll continue to rely on Spark for that. Being able to take a photo of discoloured grout in my shower and get advice for what to do about it is a killer use case.

That efficiency allowed me to spend more time with my daughters, whom I found even more charming than usual. Their creativity and spontaneous dance performances stood in sharp contrast to algorithmic systems that, for all their wonder, often offered generic, unsurprising responses. The chatbots are prediction machines, after all, based on patterns from the past.

Mayer-Schönberger said children, with their keen imaginations and constant experimentation, exemplify what sets us apart from machines. A society that becomes overly reliant on AI could ossify. “We won’t experiment as much anymore, and we will converge on the standard best option because that’s what the AI proposes,” he said.

Could these tools help us make better decisions? Possibly! When used correctly. Could they help us make decisions more quickly? Definitely. Do we want to live in a world where we are using AI to make all our decisions all the time?

The chatbots all said that was a “thought-provoking question.” I think you already know the answer.

-New York Times