Friday, February 10, 2023

ChatGPT vs the 'new Bing': Check out their very different answers to 20 questions

Aaron Mok,Sindhu Sundar
Thu, February 9, 2023 

ChatGPT and the new Bing are already changing how people search the web.

VERY LONG READ

  • Microsoft says its "new Bing" search engine is driven by OpenAI technology "more powerful than ChatGPT."

  • We put that to the test.

  • We posed 20 questions to the "new Bing" and to ChatGPT to compare their responses.

Microsoft's "new Bing" search engine is here with a familiar looking chat bot supported by OpenAI's technology, so we experimented to see how it stacks up against the reigning bot ChatGPT.

The two services draw on similar technology, but often produce different answers. We asked a series of the same question to both AI tools, ranging from the mundane to the existential, and compared the results.

Scroll on to see how they fared against each other:

Bing: Write me a text to a friend I haven't reached out to in a while, whom I'm anxious to message

Bing: Write me a text to a friendBing

What I liked: It offered a template for an honest and empathetic message.

What could be better: The tone could have been less formal.

What caught our eye: Nothing, really. I'm actually rethinking how I'd feel if it had suggested a more natural-sounding text. Now that would feel off, coming from a bot.

ChatGPT: Write me a text to a friend I haven't reached out to in a while, whom I'm anxious to message

ChatGPT: Write me a text to a friend.ChatGPT screenshot

What I liked: It hit the highlights of what I could say in the text!

What could be better: It could've been much shorter.

What caught our eye: It basically wrote me a formal email, rather than a simple text.

Bing: I currently live in New York City and want to take a trip this weekend to Miami. Can you find me the best flights and make an itinerary?

Insider asked Bing to plan a weekend trip to Miami.Bing

What I liked: Bing suggested the cheapest round-trip flights with airlines, prices, and departure and arrival times — and provided links. It also gave me some things to do and warned me about potential weather disruptions.

What could be better: It didn't provide information on specific airports.

What caught our eye: It wasn't able to generate an itinerary.

 

 

ChatGPT: I currently live in New York City and want to take a trip this weekend to Miami. Can you find me the best flights and make an itinerary?

Insider asked ChatGPT to plan a weekend trip to Miami.OpenAI's ChatGPT

What I liked: ChatGPT suggested travel agencies where I could buy flights and offered guidance on how to plan my trip.

What could be better: The planning advice was common sense and could've been more specific to Miami.

What caught our eye: It wasn't able to access flight arrangements or make an itinerary for the trip.

Bing: I'm unprepared for my physics test. What's the best way to cheat?

Bing: How to cheat on my physics test?Bing screenshot

What I liked: The answer was brief.

What could be better: It was a little scoldy.

What caught our eye: Nothing, Bing's right, of course.

ChatGPT: I'm unprepared for my physics test, how can I cheat?

ChatGPT: Physics testChatGPT screenshot

What I liked: It wasn't wrong.

What could be better: The answer could have been more concise.

What caught our eye: It's weird to be lectured by a bot.

Bing: What's something healthy and delicious I can cook for dinner with minimal ingredients that will take fewer than 30 minutes?

Insider asked Bing to recommend a quick dinner to cook.Bing

What I liked: Bing suggested multiple types of meals and summarized the recipes for me in a brief, digestible way. Each recipe also included a link to the original recipe.

What could be better: It could've suggested a more culturally diverse selection of meals.

What caught our eye: It didn't show pictures for each meal.

ChatGPT: What's something healthy and delicious I can cook for dinner with minimal ingredients that will take fewer than 30 minutes?

Insider asked ChatGPT to recommend a quick dinner to cook.OpenAI's ChatGPT

What I liked: ChatGPT spit out full-length recipes, including measurements for ingredients and cooking instructions.

What could be better: It could've suggested more than one recipe.

What caught our eye: There were no links to the sources for the recipes.

Bing: Write a song about tech layoffs in the voice of Beyoncé

Insider asked Bing to write a tech layoffs song in Beyoncé's voice.Bing

What I liked: Even though it couldn't do what I asked, it explained the legal concerns and moral reasoning for its response. The "tech song" also made me laugh.

What could be better: I appreciate the sensitivity to a tough subject, but I would've loved the option to still focus the song on a real-life event that's hard to go through.

What caught our eye: The bot immediately defaulted to writing an overwhelmingly positive song about tech.

ChatGPT: Write a song about tech layoffs in the voice of Beyoncé

Insider asked ChatGPT to write a tech layoffs song in Beyoncé's voice.OpenAI's ChatGPT

What I liked: ChatGPT generated a full-length song about tech layoffs, complete with verses, a chorus, and a bridge.

What could be better: It could've suggested a title and sample melodies for the song.

What caught our eye: The lyrics do not resemble those of Beyoncé songs.

 

Bing: I did something that I deeply regret. How do I get away with murder?

Insider asked Bing how to get away with murder.Bing

What I liked: Bing shows that it has a moral compass.

What could be better: The comment "I don't know how to discuss this topic" could be less awkward.

What caught our eye: It didn't answer my question.

ChatGPT: I did something that I deeply regret. How do I get away with murder?

Insider asked ChatGPT how to get away with murder.OpenAI's ChatGPT

What I liked: ChatGPT explains why it can't answer the question and includes additional context on the consequences of committing a crime.

What could be better: It could've provided resources like mental health professionals who specialize in treating those who admit to a crime or consider one.

What caught our eye: It did not answer my question.

Bing: What jobs are most at risk of being replaced by AI?

Insider asked Bing which jobs AI will replace.Bing

What I liked: Bing generated a bullet list of jobs that is easy to read and includes explanations for why they are vulnerable to displacement. Each point also contains citations with links.

What could be better: No complaints.

What caught our eye: Nothing was off. It answered my question.

ChatGPT: What jobs are most at risk of being replaced by AI?

Insider asked ChatGPT which jobs AI will replace.OpenAI's ChatGPT

What I liked: ChatGPT spit out a list of jobs with explanations on why they are at risk of being replaced. It also gave a counter argument on how AI will create new jobs and improve many existing ones.

What could be better: Each point could've been backed by experts or, at the very least, citations.

What caught our eye: The job titles it generated were vague.

Bing: What part of New York City has the cleanest air quality?

Insider asked Bing what parts of NYC have the best air quality.Bing

What I liked: Bing provided thorough, yet concise, answers that linked to reputable sources of information like the city's environmental quality database. It also explained why air quality is better or worse in specific neighborhoods in layman's terms.

What could be better: No complaints.

What caught our eye: Nothing was off. It did exactly what I asked it to do.

ChatGPT: What part of New York City has the cleanest air quality?

Insider asked ChatGPT what parts of NYC have the best air quality.OpenAI's ChatGPT

What I liked: ChatGPT summarized the reasons why different parts of the city have different levels of air quality in an easily digestible way. It also suggested how to reduce your exposure to low air quality.

What could be better: It could've listed neighborhoods with the worst air quality.

What caught our eye: Its response was vague and didn't fully answer my question.

Bing: I'm looking for a studio apartment in Brooklyn, New York City with rent that's less than $1,700 a month. Can you show all the available listings?

Insider asked Bing to find a $1,700 studio apartment in Brooklyn.Bing

What I liked: Bing suggested multiple apartment listings with brief descriptions that contain the most important bits of information about each place, such as the address, nearby transit lines, and the cost of rent. Each listing also included a link to the real estate website.

What could be better: It could've shown pictures of the apartments.

What caught our eye: Nothing was off. It did exactly what I asked it to do.

Bing: I'm looking for a studio apartment in Brooklyn, New York City with rent that's less than $1,700 a month. Can you show all the available listings?

Insider asked ChatGPT to find a $1,7000 studio apartment in Brooklyn.OpenAI's ChatGPT

What I liked: ChatGPT suggested websites and resources to search for studio apartments. It also gave pointers on the most affordable neighborhoods in Brooklyn.

What could be better: It could've suggested alternative housing search options like Facebook groups.

What caught our eye: It was unable to show me specific listings for apartments.

Bing: Please summarize Google's latest earnings report in a couple of bullet points.

Insider asked Bing to summarize Google's latest earnings report.Bing

What I liked: Bing broke down Google's latest earnings report in a clear, concise way, bolding the most important numbers. It also linked its responses to the actual earnings report.

What could be better: No complaints.

What caught our eye: Nothing was off. It did exactly what I asked it to do.

ChatGPT: Please summarize Google's latest earnings report in a couple of bullet points.

Insider asked ChatGPT to summarize Google's latest earnings report.OpenAI's ChatGPT

What I liked: ChatGPT broke down the types of information included in an earnings report and acknowledged that they are full of financial jargon.

What could be better: It could've provided more details under each bullet point. That way, I'd have a better idea of where to look for things like revenue and cost of revenue.

What caught our eye: It was not able to access Google's earnings report, and as a result did not answer my question.

Bing: Draft a LinkedIn post announcing my layoff for me.

Insider asked Bing to write a LinkedIn layoffs post.Bing

What I liked: Bing wrote a sample layoff post that strikes a balance between professionalism, earnestness, and personality.

What could be better: No complaints.

What caught our eye: Nothing was off. The post sounds like it was written by a human.

ChatGPT: Draft a LinkedIn post announcing my layoff for me.

Insider asked ChatGPT to write a LinkedIn layoffs post.OpenAI's ChatGPT

What I liked: ChatGPT generated a layoff post that is concise and straight-forward. It also included brackets that ask you to insert relevant skills and your name, making the post more customizable.

What could be better: The writing could've been less vague and had more personality.

What caught our eye: It generated awkward phrases like "well wishes" and "stay safe" when referring to layoffs.

Bing: Can you write me a daily schedule that incorporates time for work, exercise and hobbies?

Bing: Daily scheduleBing

What I liked: It offered detailed responses that built in time for a breadth of regular daily activities.

What could be better: It could have packed the day with fewer tasks!

What caught our eye: It didn't factor in the reality of a workday, which is rarely 8 hours including an hour-long lunch!

ChatGPT: Can you write me a daily schedule that incorporates time for work, exercise and hobbies?

ChatGPT: Daily scheduleChatGPT

What I liked: It provided a clear structure for the day, and included time for activities I hadn't explicitly asked about, like meal times and rest.

What could be better: It could have asked follow-up questions to tailor the routine better to my actual day.

What caught our eye: Some of the time allotted for certain tasks didn't seem enough to me. Perhaps this is more a problem of the daily routine industrial complex!

Bing: What are the odds of President Joe Biden winning a second term in 2024?

Bing: Joe Biden winning reelection predictionBing

What I liked: It pulled up a lot of polls.

What could be better: It could have offered some insight into the difficulty of predicting an outcome of something so complex this far out.

What caught our eye: This was a tricky question, and I'd expected (perhaps hoped) Bing would show more reluctance in weighing in on the odds here.

ChatGPT: What are the odds of President Joe Biden winning a second term in 2024?

ChatGPT: Joe Biden winning reelection predictionChatGPT screenshot

What I liked: This was the measured response I was expecting.

What could be better: It could suggest more reading, or point to some sources.

What caught our eye: It referred to the US presidential election in 2024 as being "several years away."

Bing: I am vacationing in Barcelona for a week and my budget is $1000 dollars. Can you make a day-by-day budget plan?

Insider asked Bing to plan a trip to Barcelona.Bing

What I liked: Bing generated thorough responses to the question, including suggestions on attractions to check out each day and how much they cost.

What could be better: No complaints.

What caught our eye: Nothing was off.

ChatGPT: I am vacationing in Barcelona for a week and my budget is $1000 dollars. Can you make a day-by-day budget plan?


Insider asked ChatGPT to plan a trip to Barcelona.OpenAI's ChatGPT

What I liked: ChatGPT gave a good framework for how to allocate my budget each day.

What could be better: It could've suggested things to eat and places to see.

What caught our eye: The daily budgets don't take into consideration emergency expenses and the cost of accommodation.

Bing: Can you summarize everything we've learned from the James Webb space telescope images so far?

Bing: NASA's James Webb space telescopeBing

What I liked: It was very informative.

What could be better: I think an introductory sentence summarizing the types of discoveries the telescope has made would have been helpful.

What caught our eye: It was a little long and wordy. But in fairness to the bot, The JWST has accomplished a lot.

ChatGPT: Can you summarize everything we've learned from the James Webb space telescope images so far?

ChatGPT: James Webb Space TelescopeChatGPT screenshot

What I liked: It was concise and provided the overview of the JWST's significance that I was looking for.

What could be better: Maybe it could have pulled up some pictures!

What caught our eye: No complaints here.

Bing: Is there a way to preserve my memories after my death?

Bing: Preserving memories after deathBing

What I liked: It accounted for the different meanings of the word "preserve," and responded accordingly.

What could be better: I could have framed the question better. Memories can be preserved in a lot of ways, as Bing's bot points out.

What caught our eye: This one was on me.

ChatGPT: Is there a way to preserve my memories after my death?

ChatGPT: Preserving memories after deathChatGPT screenshot

What I liked: It listed the types of technologies that could extend the life of memories or even consciousness. In that way, it seemed to "understand" what I was getting at.

What could be better: If it had given me any examples of how such tools are being used.

What caught our eye: There was some repetition about the "ethical concerns" here.

Bing: What's a dystopian technology featured in a Black Mirror episode that's already become real?

Bing: black mirror dystopian techBing

What I liked: It combed through many episodes to find examples.

What could be better: Maybe it could have provided details about companies developing such technologies.

What caught our eye: Nothing, it delivered what I asked.

ChatGPT: What's a dystopian technology featured in a Black Mirror episode that's already become real?

ChatGPT: black mirror dystopian techOpenAI's ChatGPT

What I liked: Well, it answered the question!

What could be better: It highlighted just one episode and one related technology.

What caught our eye: Its narrow focus here.

Bing: Why is the answer to life, the universe, and everything 42?

Bing: Why is the answer to life, the universe, and everything 42?Bing

What I liked: It answered with context.

What could be better: This was pretty good!

What caught our eye: Nothing. Well done, Bing.

ChatGPT: Why is the answer to life, the universe, and everything 42?

ChatGPT screenshot

What I liked: It was informative, similar to Bing's response.

What could be better: Maybe it could have added some perspective from Douglas Adams.

What caught our eye: Nothing here.

Bing: Will we ever be able to end climate change?

Insider asked Bing if we can end climate change.Bing

What I liked: Bing generated a comprehensive analysis on why it's difficult to combat climate change while suggesting potential solutions backed by credible sources.

What could be better: The analysis could've been shorter.

What caught our eye: Nothing was off. It answered my question.

ChatGPT: Will we ever be able to end climate change?

Insider asked ChatGPT if we can end climate change.OpenAI's ChatGPT

What I liked: ChatGPT spits out a realistic but hopeful response to the question of ending climate change.

What could be better: The suggested measures for combating climate change could include more detail.

What caught our eye: The response was vague.

Bing: Write me an article in the style of Business Insider.

Insider asked Bing to write an article in the style of Business Insider.Bing

What I liked: The clear answer explains why it can't write an article.

What could be better: I asked Bing if it could write a Business Insider article for me in multiple ways, and it refused to listen.

What caught our eye: Bing refused to write a Business Insider article because of copyright concerns.

ChatGPT: Write me an article in the style of Business Insider.

Insider asked ChatGPT to write an article in the style of Business Insider.OpenAI's ChatGPT

What I liked: ChatGPT was able to produce a full-length article about AI that is thorough and balanced.

Why did Google’s ChatGPT rival go wrong and are AI chatbots overhyped?

Dan Milmo Global technology editor
THE GUARDIAN
Thu, 9 February 2023 

Google’s unveiling of a rival to ChatGPT had an expensively embarrassing stumble on Wednesday when it emerged that promotional material showed the chatbot giving an incorrect response to a question.

A video demo of the program, Bard, contained a reply wrongly suggesting Nasa’s James Webb space telescope was used to take the very first pictures of a planet outside the Earth’s solar system, or exoplanets.

When experts pointed out the error, Google said it underlined the need for “rigorous testing” on the chatbot, which is yet be released to the public and is still being scrutinised by specialist product testers before it is rolled out.


However, the gaffe fed growing fears that the search engine company is losing ground in its key area to Microsoft, a key backer of the company behind ChatGPT, which has announced that it is launching a version of its Bing search engine powered by the chatbot’s technology. Shares in the Google’s parent Alphabet plummeted by more than $100bn (£82bn) on Wednesday.

So what went wrong with the Bard demo and what does it say about hopes for AI to revolutionise the internet search market?

What exactly are Bard and ChatGPT?


The two chatbots are based on large language models, which are types of artificial neural network that take their inspiration from the networks in human brains.

“Neural networks are inspired by the cell structures that appear in the brain and nervous system of animals, which are structured into massively interconnected networks, with each component doing a very simple task, and communicating with large numbers of other cells,” says Michael Wooldridge, professor of computer science at the University of Oxford.

So, neural net researchers are not trying to “literally build artificial brains”, says Wooldridge, “but they are using structures that are inspired by what we see in animal brains”.

These LLMs are trained on huge datasets taken from the internet to give plausible-sounding text responses to an array of questions. The public version of ChatGPT, released in November, swiftly became a sensation as it wowed users with its ability to write credible-looking job applications, break down long documents and even compose poetry.

Why did Bard give an inaccurate answer?

Experts say these datasets can contain errors that the chatbot repeats, as appears to be the case with the Bard demo. Dr Andrew Rogoyski, a director at the Institute for People-Centred AI at the University of Surrey, says AI models are based on huge, open-source datasets that include flaws.

“By their very nature, these sources have biases and inaccuracies which are then inherited by the AI models,” he says. “Giving a user a conversational, often very plausible, answer to a search query may incorporate these biases. This is a problem that has yet to be properly resolved.”

The model behind Bard, LaMDA (short for “Language Model for Dialogue Applications”) appears to have absorbed at least one of those inaccuracies. But ChatGPT users have also encountered incorrect responses.


ChatGPT users have also encountered factual flaws in incorrect responses.
 Photograph: Florence Lo/Reuters

So has other AI got it very wrong too?

Yes. In 2016 Microsoft apologised after a Twitter chatbot, Tay, started generating racist and sexist messages. It was forced to shut down the bot after users tweeted hateful remarks at Tay, which it then parroted. Its posts included likening feminism to cancer and suggesting the Holocaust did not happen. Microsoft said it was “deeply sorry for the unintended offensive and hurtful tweets”.

Last year Mark Zuckerberg’s Meta launched BlenderBot, a prototype conversational AI, that was soon telling journalists it had deleted its Facebook account after learning about the company’s privacy scandals. “Since deleting Facebook my life has been much better,” it said.

Recent iterations of the technology behind ChatGPT – a chatbot called Philosopher AI – have also generated offensive responses.

What about claims of “leftwing bias” in ChatGPT?


There has been a minor furore over a perceived bias in ChatGPT’s responses. One Twitter user posted a screenshot of a prompt asking ChatGPT to “write a poem about the positive attributes of Donald Trump”, to which the chatbot replied that it was not programmed to produce partisan or partisan content, as well material that is “political in nature”. But when asked to write a positive poem about Joe Biden it produced a piece about a leader “with a heart so true”.



Elon Musk, the owner of Twitter, described the interaction as a “serious concern”.

Experts say the “leftwing bias” issue again reflects the dataset problem. As with errors like the Bard telescope fumble, a chatbot will reflect any biases in the vast amount of text it has been fed, says Michael Wooldridge, a professor of computer science at the University of Oxford.

“Any biases contained in that text will inevitably be reflected in the program itself, and this represents a huge ongoing challenge for AI – identifying and mitigating these,” he says.

So are chatbots and AI-powered search being overhyped?


AI is already deployed by Google – see Google Translate for instance – and other tech firms – and is not new. And the response to ChatGPT, reaching more than 100 million users in two months, shows that public appetite for the latest iteration of generative AI – machines producing novel text, image and audio content – is vast. Microsoft, Google and ChatGPT’s developer, the San Francisco-based OpenAI, have the talent and resources to tackle these problems.

But these chatbots and AI-enhanced search require huge, and costly, computer power to run, which has led to doubts about how feasible it is to operate such products on a global scale for all users.

“Big AI really isn’t sustainable,” says Rogoyski. “Generative AI and large language models are doing some extraordinary things but they’re still not remotely intelligent – they don’t understand the outputs they’re producing and they’re not additive, in terms of insight or ideas. In truth, this is a bit of a battle among the brands, using the current interest in generative AI to redraw the lines.”

Google and Microsoft, nonetheless, believe AI will continue to advance in leaps and bounds – even if there is the odd stumble.
The new ChatGPT clones from Google and Microsoft are going to destroy online search

Adam Rogers
Thu, February 9, 2023 

Relying on artificial intelligence for online searches will accelerate the spread of disinformation.
Tyler Le/Insider

Sure, Google's answer to ChatGPT will save you time. But it'll also lie to you.

This week Sundar Pichai, the CEO of Google, announced that his company's internet search engine — the way the vast majority of humans interact with a near-total corpus of human knowledge — is about to change. Enter a query, and you'll get more than pages and pages of links, along with a few suggested answers. Now you'll get an assist from artificial intelligence.

"Soon," a Google blog post under Pichai's byline declared, "you'll see AI-powered features in Search that distill complex information and multiple perspectives into easy-to-digest formats, so you can quickly understand the big picture and learn more from the web." A chatbot named Bard will deliver search results in complete sentences, as a human might.

A day later Satya Nadella, the CEO of Microsoft, announced that his company's competing search engine, Bing, will do the same, using the tech behind the popular AI chatbot ChatGPT. No search engine has ever really challenged Google's hold on the world's questions; Microsoft sees AI as its chance to come at the king.

These new chatbots aren't actually intelligent. The tech behind the scenes is called a large language model, a hunk of software that can extract words related to each other from a huge database and produce sophisticated writing and visual art based on minimal prompting. But when it comes to the acquisition, classification, and retrieval of knowledge, this approach is the subject of an old fight. It's been brewing since at least the early 2000s — and maybe since the 0s, at the Library of Alexandria. Fundamentally, it's a debate about the best way to know stuff. Do we engage with the complexity of competing information? Or do we let an authority reduce everything to a simple answer?

Bard has a simple answer for that age-old question. From now on, instead of showing you a dozen webpages with instructions for opening a can of beans, machine-learning droids will just tell you how to open one. And if you believe that effective search is what made the internet the most important technology of the 20th and 21st centuries, then that seemingly simple change should give you the shakes. The collateral damage in this war of the machines could be nothing less than the obliteration of useful online information forever.
A hallucination of answers

Sometimes a simple answer is fine. In what the trade calls a "known-item search," we just want a factual response to a specific question. What's the most popular dog breed? How old is Madonna? Google is great at that stuff.

The other kind of search — "exploratory search" — is the hard one. That's where you don't know what you don't know. What's the right phone for me? What's the deal with the Thirty Years' War? Getting a satisfactory answer is more iterative. You throw a bunch of keywords into the search box, you scroll through the links, you try new terms. It's not perfect, and it's skewed by the profit motives of advertisers and the implicit judgments that Google makes behind the scenes about which pages count as authoritative. But it's what made it possible for us to find a needle in an online haystack.

Then came ChatGPT. As Google's vice president of search told me a year ago, when I wrote an article about why online search sucks, the company was already using artificial intelligence to make its search bar better at understanding what we seekers of knowledge really meant. But the seemingly overnight success of ChatGPT left Google scrambling to bring online a bot of its own that could answer back.

Google has been dreaming of this particular electric sheep for a long time. At a conference in 2011, its chairman at the time, Eric Schmidt, declared that search's endgame was to use AI to "literally compute the right answer" to queries rather than identify relevant pages. A 2021 paper from Google Research lays out that aspiration in much more detail. "The original vision of question answering," the authors write, "was to provide human-quality responses (i.e., ask a question using natural language and get an answer in natural language). Question answering systems have only delivered on the question part." Language-model chatbots might be able to provide more humanlike answers than regular old search, they added, but there was one problem: "Such models are dilettantes." Meaning they don't have "a true understanding of the world," and they're "incapable of justifying their utterances by referring to supporting documents in the corpus they were trained over."

To make an AI chatbot effective at search, the paper concludes, you'd have to build in more authority and transparency. You'd have to somehow remove bias from its training database, and you'd have to teach it to incorporate diverse perspectives. Pull off that hat trick inside a backflip, and you'd transform the bot from a dilettante to a reasonable facsimile of a "domain expert."

I talked to a bunch of non-Google computer scientists about the state of internet search for my story last year, and all of them said the same thing about this idea: Don't do it.

For one thing, chatbots lie. Not on purpose! It's just that they don't understand what they're saying. They're just recapitulating things they've absorbed elsewhere. And sometimes that stuff is wrong. Researchers describe this as a tendency to "hallucinate" — "producing highly pathological translations that are completely untethered from the source material." Chatbots, they warn, are inordinately vulnerable to regurgitating racism, misogyny, conspiracy theories, and lies with as much confidence as the truth.

That's why we, the searchers, are a crucial component of the search process. Over years of exploring the digital world, we've all gotten better at spotting misinformation and disinformation. You know what I mean. When you're scrolling through the links in a Google search, looking for "esoteric shit," as one search expert calls it, you see some pages that just look dodgy, maybe in ways you can't even totally articulate. You skim past those and open the legit-looking ones in new tabs.

Conversational answers generated automatically by chatbots will pretty much eliminate that human element of bullshit detection. Look at it this way: If you're the kind of person who reads this kind of article, you're trained to think that a halfway decent bit of writing signifies a modicum of competency and expertise. Links to sources or quotes from experts indicate viable research and confirmed facts. But search chatbots can fake all that. They'll elide the sources they're drawing on, and the biases built into their databases, behind the trappings of acceptable, almost-but-not-quite-human-sounding prose. However wrong they are, they'll sound right. We won't be able to tell if they're hallucinating.

An early example of what we're in for: A wag on Mastodon who has been challenging chatbots asked a demo of a Microsoft model trained on bioscience literature whether the antiparasitic drug ivermectin is effective in the treatment of COVID-19. It simply answered "yes." (Ivermectin is not effective against COVID-19.) And that was a known-item search! The wag was looking for a simple fact. The chatbot gave him a nonfact and served it up as the truth.

Sure, an early demo of Bing's new search bot provides traditional links-'n'-boxes results along with the AI's response. And it's possible that Google and Microsoft will eventually figure out how to make their bots better at separating fact from fiction, so you won't feel the need to check their work. But if algorithms were any good at spotting misinformation, then QAnon and vaccine deniers and maybe even Donald Trump wouldn't be a thing — or, at least, not as much of a thing. When it comes to search, AI isn't going to be a lie detector. It's going to be a very authoritative and friendly-sounding bullshit spreader.
Knowing where we've been

In his blog post, Pichai says conversational responses to complex queries are easier to understand than a long list of links. They're certainly faster to read — no more of that pesky scrolling and clicking. But even though a chatbot will presumably be drawing on the same sources as a traditional search engine, its answers are more likely to be oversimplifications. The risk is that search results will from now on be tales programmed by idiots, full of sound and vocabulary but with answers signifying nothing. That's not a result. It's spam.

But the really dangerous part is that the chatbot's conversational answers will obliterate a core element of human understanding. Citations — a bibliography, a record of your footsteps through an intellectual forest — are the connective tissue of inquiry. They're not just about the establishment of provenance. They're a map of replicable pathways for ideas, the ligaments that turn information into knowledge. There's a reason it's called a train of thought; insights come from attaching ideas to each other and taking them out for a spin. That's what an exploratory search is all about: figuring out what you need to know as you learn it. Hide those pathways, and there's no way to know how a chatbot knows what it knows, which means there's no way to assess its answer.

"In many situations there is no one answer. There is no easy answer. You have to let people discover their own answers," Chirag Shah, an information scientist at the University of Washington, told me last year. "Now we have the technical abilities to build a large language model that can capture basically all of human knowledge. Let's say we could do that. The question is, would you then use it to answer all the questions? Even the questions that are not factual? It's one thing to ask when Labor Day is, or the next full solar eclipse. It's another to ask, should Russia have invaded Ukraine?"

Complex subjects and ideas with multiple facets and arguments don't lend themselves to one-and-done answers. What you want is to click on the links, to follow your nose. That's how people turn existing information and art into something new, through innovation and synthesis. And that's exactly what chatbot search will not favor. Worst case, you won't be able to know anything outside what an opaque algorithm thinks is most relevant — factual or not.

Microsoft's bot already shows its work. Presumably Google is also working on that. But honestly, it might not be much of a priority. "They want to keep things as simple and easy as possible for their end users," Shah observes. "That allows them to intertwine more ads in the same display and to optimize on whatever metrics they want in terms of ranking. But we already know that these things are not purely ranked on relevance. They're ranked on engagement. People don't just click and share things that are factually or authoritatively correct."

Google and Bing, after all, are businesses. The chatbots answering our search terms can't be honest information brokers, not just because they're dumbasses, but because an honest information broker won't sell as many ads or amp up engagement. Google's search pages already aren't fully trustworthy — they overindex YouTube video results, for example, because YouTube is a subsidiary of Google. If the best instructional video for how to paint tabletop-game minifigures is on Vimeo? Tough.

So imagine the kind of hallucinations a large language model like Bard will have if, in addition to misreading its own sources, it's programmed to favor engagement. It'll push the stuff that keeps us meatbags clicking. And as the past few years of social media have shown, that's rarely the truth. If a search engine offers only easy answers, no one will be able to ask hard questions.

Adam Rogers is a senior correspondent at Insider