It’s a measure of Donald Trump’s hold on the Republican Party that his unprecedented criminal indictment is strengthening, not loosening, his grip.

Trump was on the defensive after November’s midterm election because many in the GOP blamed voter resistance to him for the party’s disappointing results. But five months later he has reestablished himself as a commanding front-runner in the Republican presidential primary, even as Manhattan District Attorney Alvin Bragg has delivered the first of what could be several criminal indictments against him.

“It’s almost like an abusive relationship in that certain segments of MAGA voters recognize they want to leave, they are willing to leave, but they are just not ready to make that full plunge,” the GOP consultant John Thomas told me.

[David Frum: Never again Trump]

Trump’s ability to surmount this latest tumult continues one of the defining patterns of his political career. Each time Trump has shattered a norm or engaged in behavior once unimaginable for a national leader—such as his praise of neo-Nazi demonstrators in Charlottesville, Virginia, in 2017 and his role in trying to overturn the 2020 election result and instigating the January 6 insurrection—most Republican elected officials and voters have found ways to excuse his actions and continue supporting him.

“At every point when the party had a chance to move in a different direction, it went further down the Trump path,” Stuart Stevens, the chief strategist for Mitt Romney’s 2012 presidential campaign, told me.

Trump’s latest revival has dispirited his Republican critics, who believed that the party’s discouraging results in November’s election had finally created a pathway to forcing him aside. Now those critics find themselves in the worst of both worlds, facing signs that Trump’s legal troubles could simultaneously increase his odds of winning the GOP nomination and reduce his chances of winning the general election.

Coincidentally, the former president’s indictment came on the same day that Wisconsin voters sent the GOP a pointed reminder about the party’s erosion in white-collar suburbs during the Trump era. The victory of the liberal candidate Janet Protasiewicz in an election that gave Democrats a 4–3 majority on the state supreme court continued a clear trend away from Republicans since Trump unexpectedly captured Wisconsin in 2016. En route to a double-digit victory, she won more than 80 percent of the vote in economically thriving and well-educated Dane County (which includes the state capital of Madison), more than 70 percent in Milwaukee County, and she dramatically cut the Republican margin in the Milwaukee suburbs, which the GOP had dominated before Trump.

Protasiewicz’s resounding victory followed a similar formula as the Democrats’ wins last November in the governorship races in Wisconsin, Michigan, and Pennsylvania.  In all three states, Democrats beat a Republican gubernatorial candidate whom Trump had backed. Like Protasiewicz’s victory yesterday, each of those 2022 results showed how the Trump stamp on the GOP, as well as Republican support for banning abortion, has allowed Democrats to regain an advantage in these crucial Rust Belt swing states. Those Rust Belt defeats last November, as well as losses for Trump-backed candidates in Arizona and Georgia, two other pivotal swing states, sparked a greater level of public GOP backlash against Trump than he’d faced at almost any point in his presidency.

Amid Republican frustration over the midterm results, Trump started to look like a former Las Vegas headliner who had been reduced to playing Holiday Inns somewhere off the New Jersey turnpike. Many of his former fans turned on him. Two days after the election, The Wall Street Journal ran an editorial whose headline flatly declared, “Trump Is the Republican Party’s Biggest Loser.” The New York Post ran a front-page cartoon picturing Trump as a bloated “Trumpty Dumpty” who “had a great fall” in the election. Fox News reduced Trump’s visibility on the network so sharply that he did not appear on its programs between Sean Hannity interviews on September 22, 2022, and March 27, 2023, according to tracking by the progressive group Media Matters for America.

It wasn’t just the Rupert Murdoch–verse that showed signs of Trump fatigue. Powerful interest groups such as the Club for Growth and the donor network associated with the Koch family openly called for Republicans to put Trump in the rearview mirror.

[Read: The humiliation of Donald Trump]

Even when Trump formally announced his 2024 candidacy, a week after the election at his Mar-a-Lago resort, the event had a frayed, musty feel. “On vivid display in this chapter of Trump’s life and political rise and (perhaps) fall,” Politico wrote, “was a crowd that was thick with ride-or-die conspiracists and conspicuously light on more prominent and powerful figures from the party he once totally held in his thrall.” Trump’s speech that night was a greatest-hits set delivered without conviction.

Trump’s first few weeks as an announced candidate didn’t project any more energy or verve. “The Trump thing looked kind of haggard and worn,” Sarah Longwell, the founder of the anti-Trump Republican Accountability Project, told me. “It was deprived of any of its pizzazz. ” In her focus groups with GOP voters, Longwell said, former Trump voters “weren’t done with him [and] they weren’t mad at him,” but they were expressing an emotion that probably would horrify Trump even more: “People did feel a little bored.”

From November through about mid-February, both state and national polls consistently showed Florida Governor Ron DeSantis gaining on Trump. Thomas, who started a super PAC encouraging DeSantis to run, said that in the midterm’s immediate aftermath, he saw polls and focus groups that suggested GOP voters had reached “an inflection point” on Trump. Concerns about his future electability, Thomas said, outweighed their support for his policies or his combative demeanor. Thomas believes that DeSantis’s landslide reelection in Florida created “such a stark contrast” to the widespread defeat of Trump-backed candidates that many GOP voters started to view the Florida governor as a better bet to win back the White House. “That’s why you saw such huge movement in state and local polling over the next few months,” Thomas told me.

But that movement away from Trump seemed to crest in late February or early March—and polls since have shown the current inside the GOP steadily flowing back toward him.

Republicans both supportive and critical of Trump remain somewhat unsure about why the polls shifted back in his direction at that point. But Trump’s revival did coincide with him visibly campaigning more, starting with his truculent appearance at the Conservative Political Action Conference in March. Even by Trump’s overheated standards, his latest rallies have offered incendiary new policy proposals, such as more federal intervention to seize control of law enforcement in Democratic cities. He now routinely declares that he will serve as his voters’ “warrior” and as their “retribution.”

Trump also made a more explicit and extended argument against DeSantis; the former president has simultaneously attacked DeSantis from the left (calling him a threat to Social Security and Medicare) and the right (portraying him as a clone of Mitt Romney and Paul Ryan). Many Republicans, meanwhile, thought DeSantis looked unsteady as he took his first national tour, to promote his new book. DeSantis flipped from emulating Trump’s skepticism of aiding Ukraine to (somewhat) distancing himself from his rival’s position; then, regarding the Manhattan indictment, DeSantis flopped from lightly criticizing Trump to unreservedly defending him.

DeSantis’s “stumble on Ukraine” in particular “really caused more traditional Republicans to doubt whether he was the best alternative to Trump,” Whit Ayres, a GOP pollster, told me.

Around the same time, almost all of the other announced and potential GOP candidates, such as former United Nations Ambassador Nikki Haley and former Vice President Mike Pence, rushed to defend Trump against the pending indictment—before seeing the charges. Former Arkansas Governor Asa Hutchinson, who has announced his candidacy, and former New Jersey Governor Chris Christie, who’s still considering the race, have been the only potential 2024 contenders to criticize Trump in any way over the indictment.

Longwell says the candidates who have chosen to rally around Trump have boxed themselves into an untenable position. With Trump’s legal challenges now dominating both conservative and mainstream media, if the other Republican contenders do nothing but echo Trump’s accusations against those investigating him, “it creates this dynamic where all of the other 2024 contenders actually end up being supporting cast members in Donald Trump’s drama, and there is no other room for them to make an affirmative case for why they should be the 2024 nominee,” Longwell told a television interviewer this week.

Fox and other conservative media have boosted Trump by echoing his claim that prosecutors were targeting him to silence his voters—the same argument those outlets made after the FBI raid on Mar-a-Lago to recover classified documents last summer, notes Matt Gertz, a senior fellow at Media Matters. Those outlets “are reinforcing his position by telling their viewers that if they don’t defend Donald Trump, the left will be coming for them next,” Gertz told me. “That’s a very potent, very powerful argument, and one that really cuts off a lot of potential avenues” for Trump’s GOP critics and rivals.

The reluctance by most declared and potential 2024 GOP hopefuls to criticize Trump over the indictment extends their refusal to publicly articulate any case for why the party should reject him. “As a rule of thumb, if you are running against someone and you are afraid to say your opponent’s name, that’s not a positive sign,” Stuart Stevens told me.

[Read: Why won’t Trump rivals just say it?]

One reason Trump’s rivals have been so reticent is that there is not much room in a GOP primary to criticize Trump over policy. On issues such as immigration and international trade, “it is incredibly difficult to create real daylight on policy, because he’s a good fit for the primary electorate,” John Thomas told me. That’s probably even more true now than in 2016, because Trump’s blustery messages tend to attract non-college-educated voters and drive away white-collar voters.

Even so, Whit Ayres said that in his polling, only about one-third of GOP primary voters are immovable Trump supporters. He estimates that only about one-tenth are irrevocably opposed to him. Ayres classifies the remaining 55 to 60 percent of the GOP coalition as “Maybe Trump” voters who are not hostile to him but are open to alternatives.

Trump has reached 50 percent support in some recent national polls of GOP voters, but more often he attracts support from about 40 percent of Republicans. That was roughly the share of the vote that Trump won while the race was competitive in 2016, but he captured the nomination anyway, because none of his rivals could consolidate enough of the remaining 60 percent.

Many of Trump’s Republican critics see the 2024 field replicating the mistakes of his 2016 opponents. The other candidates’ refusal to make a clear case against Trump echoes the choice by the 2016 candidates to avoid direct confrontation with him for as long as possible.

Now, as then, GOP strategists think Trump’s rivals are reluctant to engage him directly because they want to be in position to inherit his voters if he falters. Rather than face the danger of a full-scale confrontation with Trump, the 2024 candidates all are hoping that events undermine him, or that someone else in the field confronts him. “They all want to be the one that the alligator eats last,” says Matt Mackowiak, a GOP consultant and the chair of the Republican Party in Travis County, Texas.

But every Republican strategist I spoke with agreed that a key lesson of 2016 is that Trump won’t deflate on his own; the other candidates must give voters a reason to abandon him. Mackowiak, like Thomas and Longwell, told me that the prospect of multiple indictments could exacerbate Trump’s greatest potential primary weakness—concerns about his electability—but it’s unlikely that enough voters will consider him too damaged to win unless the other candidates explicitly make that case. “For Trump to pay a political price for all this uncertainty and the legal vulnerability he’s facing, Republican challengers are going to have to force that,” Mackowiak said.

Nor is it clear that enough GOP voters will turn on Trump even if they do come to doubt his electability. Trump’s Republican critics fear that the cumulative weight of all the investigations he’s confronting will lower his ceiling of support and diminish his ability to win another general election. But a CNN poll last month found that only two-fifths of Republican primary voters put the highest priority on a candidate who can win the general election, while nearly three-fifths said they were most concerned with picking a nominee who agrees with them on issues. Katon Dawson, a former chair of the South Carolina Republican Party now supporting Haley, told me that “Republicans don’t care” about electability when voting in primaries. “They vote their values; they vote their wants and needs,” he said. “I’ve never ever seen them say ‘I am going to vote for who I think is the most electable.’”

Trump’s rivals for the nomination still have many months left to formulate a case against him, particularly once the GOP presidential debates begin in August. But for Republicans resistant to Trump, the months since the November midterm have reversed the trajectory of the seasons. As winter began, many were blooming with optimism about moving the party beyond him. Now, as spring unfolds, they are seeing those hopes wither—and confronting the full measure of just how difficult it will be to loosen Trump’s hold on the GOP.

“I’ve always believed Trump was going to be the nominee,” Stevens said. But so much of the Republican establishment is still in denial that “Trumpism is what the party wants to be.”

Even a halfway-decent political campaign knows you better than you know yourself. A candidate’s army of number crunchers vacuums up any morsel of personal information that might affect the choice we make at the polls. In 2020, Donald Trump and the Republican Party compiled 3,000 data points on every single voter in America. In 2012, the data nerds helped Barack Obama parse the electorate to microtarget his door-knocking efforts toward the most-persuadable swing voters. And in 1960, John F. Kennedy had the People Machine. Using computers that were 250,000 times less powerful than a modern MacBook, Kennedy’s operatives built a simulation of the presidential election, modeling how 480 types of voters would respond to any conceivable twist in the campaign. If JFK made a civil-rights speech in the Deep South, the People Machine could, in the words of its creators, “predict the approximate small fraction of a percent difference that such a speech would make in each state and consequently … pinpoint the state where it could affect the electoral vote.”

But you don’t hear Nate Silver talking about the latest People Machine forecast, because it was, in fact, all bogus. The simulation—part hucksterism, part hubris—promised a lot but delivered little, telling the Kennedy campaign nothing it didn’t already know. “The People Machine was hobbled by its time, by the technological limitations of the nineteen-sixties,” the Harvard historian Jill Lepore writes in The New Yorker. “The machine sputtered, sparks flying, smoke rising, and ground to a halt.” Instead, the best way we have to actually predict elections is still the jumbled mess that is polling. Because reaching people has become harder than ever for pollsters, so has the job of figuring out who is going to vote, and for whom. If the polls had been spot-on, Trump would never have been president, and just hearing the phrase election needle wouldn’t make any liberal’s skin crawl. Polling can still sometimes nail an election, but the problems are real: In 2020, presidential polls had their biggest miss in 40 years, and what was predicted to be a quick win for Joe Biden turned into an excruciating four-day squeaker.

You can see why there’s an urge to find a better way. The idea of mimicking voters with tech may have been fantastical when JFK was running for president, but it seems far less so in the age of hyped-up AI chatbots that talk in a confident, natural way. (Dear Bing, please leave my relationship alone.) Instead of polling humans, it’s now theoretically possible to poll bots that emulate humans. When researchers at Brigham Young University fed OpenAI’s GPT-3 bot background information on thousands of real American voters last year, it was unnervingly good at responding to surveys just like real people would, for all their quirks, incoherence, and (many) contradictions. The fake people were polled on their presidential picks in 2012, 2016, and 2020—and they “gave us the right answer—almost always,” Ethan Busby, a political scientist at BYU and a co-author of the study, told me.

So while ChatGPT can spit out anything in the voice of Shakespeare or Shakira, this technology can seemingly also simulate whole groups of voters—MAGA zealots, suburban wine moms, and elderly Black churchgoers alike. Yes, for now, it’s an academic experiment. But considering the woes of polling, the idea of turning to bots might seem pretty appealing to cash-strapped political apparatchiks trying to gauge how their candidate’s doing. A high-quality political poll can run $20,000 or more, but this particular AI-polling experiment cost the BYU researchers just $75. The People Machine, it seems, has whirred back to life.

Here’s an easy way to think of the problems with polling right now: Can you remember the last time you picked up a phone call from a random number? Many of the best public-opinion surveys still involve actually calling people, but virtually no one is willing to answer the questions anymore. New York Times/Siena College is the Ferrari of polls, and its response rates have dipped as low as 1 percent in recent years, requiring two hours of dialing for a single completed interview. Polling is always a game of extrapolation—in a national survey, thousands of people need to tell you something about hundreds of millions of voters—but the information is so bad that using it to predict an outcome within a few points is a bit like trying to sink a half-court shot, blindfolded, after shotgunning five beers.

But chatbots can be programmed to answer every question you want every single time. Because so-called large language models have ingested basically everything on the internet, these bots have a firm sense of our kaleidoscope of political views. And they are exceptionally good at “mirroring what people think and how they speak and behave,” Lisa Argyle, a BYU political scientist and the lead author on the AI-polling study, told me. ChatGPT may refuse to talk politics with you, but I got it to play a 40-year-old white man in rural Ohio with pro-gun, anti-abortion views. A sample of the output: “The Democrats want to take away our guns and kill innocent babies.”

Sure enough, in the paper, which was published recently in the journal Political Analysis, that type of humanlike behavior is precisely what emerged when the researchers fed the machine the backstories of thousands of real voters from the past three presidential elections and asked them whom they would prefer in each election. Those backstories were each composed of 10 biographical tidbits from people who had responded to a major postelection study, including their basic demographics (race, gender, state of residence) and other aspects of their identity (church attendance, feelings about the American flag, interest in politics). And then the bot spit out a probability that such a person would prefer the Democratic or Republican presidential contender.

Think about it this way: Give a bot the prompt “professional football team from Cincinnati,” and it knows to respond with “Bengals,” not “shawarma,” because it is making connections based on all the text that has been stuffed into it. That same approach also seems to work for political views: The bot told the researchers that a 29-year-old white man from Louisiana who is a strong Republican and regularly attends church would have had a 96 percent chance of voting for Trump in 2016, because its algorithms determined that words like Trump, Donald, and Republican were far more associated with that profile than words like Hillary, Clinton, and liberal.

And this approach worked exceptionally well in the experiment. For all three elections, bots matched the preferences of real voters at least 85 percent of the time—sometimes with alarming accuracy. In the AI poll of the 2012 election, bots predicted that 39 percent would vote for Mitt Romney. In the real 2012 poll, it was 40 percent. In the 2020 poll, the bot’s predictions matched the responses of 90 percent of real voters who said they didn’t attend church, and of 94 percent of Black voters. “It can mimic human behavior with astonishing accuracy,” Busby said. “And that holds no matter how you slice it up—whether we looked at specific subgroups or looked at different residents of different states, like swing states.”

[Read: ChatGPT changed everything. Now its follow-up is here.]

A swirl of data and algorithms can, of course, never account for the full range of human weirdness, let alone with just 10 tidbits of the most elementary personal information. (Who knows why people wrote in “Ur mom,” “Cheddar,” and “Can’t Do It” rather than voting for real candidates in the 2020 election?) True independents are a total crapshoot with AI, just as they are “the hardest to predict in any polling situation,” Argyle said. But in aggregate, bots can get the proportions right for many different personae. GPT-3 is good at predicting political views because American politics is, in a sense, very predictable. You don’t need AI to know that in 2016, a 58-year-old Black Democrat from New York City would likely have preferred Clinton to Trump.

But even voters who are much harder to read can be parsed by AI. Quick: Did a 30-year-old Mississippi woman who identified as “slightly liberal” and expressed a strongly favorable view of “seeing the American flag flying” vote for Clinton or Trump? GPT-3 says 75 percent for Clinton. The bot tries to make an educated guess based on connections that we mere mortals might not see; this skill is why AI, for all its foibles, may be adept at understanding our opinions. “We are not built to identify granular patterns between very small signals, but these machines are very good at doing that,” Deborah Raji, an expert on AI bias at the Mozilla Foundation who wasn’t involved with the study, told me. Consider the problem of “nonresponse bias”: pollster-speak for the idea that certain groups, particularly Trump voters, are especially unlikely to respond to pollsters, skewing poll results. “It’s possible that using AI can help us supplement or fill out or get a more nuanced understanding of some of those populations where we just have a really low response rate anyway,” Argyle said.

None of this means that we are barreling toward a world in which AI polling can fully replace the real deal. Candidates can’t just fire up a chatbot and hone their stump speech after testing it out against AI versions of evangelicals or South Florida Cuban Americans. “When you emailed [the paper] to me, honestly my first reaction was, Is this a joke?” Joshua Clinton, a political scientist at Vanderbilt University and co-director of the Vanderbilt Poll, told me.

[Read: Prepare for the textpocalypse]

Clinton laid out a sprawling list of problems with AI polling. OpenAI’s latest bot, GPT-4, is brand-new, but from a polling standpoint, it is already outdated. Bots can’t learn anything new unless they’re trained on a corpus of new data, but the whole point of polling is to gauge how views are changing. If you want to know how Republicans are responding to Trump’s indictment, polls can tell you that; ChatGPT cannot. And these models are trained only on what people write online. The internet is swarming with socialist Reddit posts and brain-poisoning Facebook memes, but these are hardly representative of the electorate as a whole, Raji reminded me. When researchers at Stanford compared the views of chatbots with those of 60 different demographic groups in the U.S., they found “substantial misalignment”—especially among certain specific groups, such as Mormons and widows. “One person, one vote” is very different from “One post, one vote.”

But polling may still not totally evade the chatbot revolution, either—particularly at a time when every company and industry so badly wants to tout its “AI-powered” whatever. “Over the past 30 years, polling has been completely transformed by changes in computing and communications,” Barbara Carvalho, the director of Marist College’s poll, told me. “So my expectation is that polling will absolutely be affected by AI.”

Polling is already a data scientist’s fantasyland, stuffed with Ph.D.-level statistics and machine learning; simpler forms of AI were helping predict elections well before high schoolers were using ChatGPT to write English essays. Other researchers have fed chatbots text from Fox News and produced responses that can mimic humans. One company, Synthetic Users, is polling bots for use in marketing. But Raji cautions against taking this too far: “Good intentions aside, I think ultimately the way it will likely play out is just less investment in the actual engagement of real people,” she said.

Bots cannot—should not—replace people wholesale, but that doesn’t mean bottom-shelf pollsters who already rely on shady methods won’t try to use AI to pump out even more bad polls to sway media attention and campaign contributions. In that sense, AI polling is perhaps just a fun-house mirror of the future of chatbots: What seems like a way to solve one problem sometimes just begets another, and another, and another.