Every year my family gathers for a vacation. Just the four of us: my mom, my dad, my sister and me. This time, however, my parents brought a fifth member to our New England getaway: Claude, Anthropic's AI chatbot. Or, as my mom calls it, "The nice man who answers all my questions."

My parents, lawyers in their 70s, live in Australia. My dad is pretty hip to tech, but my mom has always been a proud Luddite. She'd sooner use a carrier pigeon than Zoom and couldn't care less about social media ("I don't need to see photos of every flower in my friends' gardens").

Two weeks before our vacation, however, she heard her company was strongly encouraging employees to use AI. "Quick," she said to a colleague, "show me how to use Claude."

As we gathered for breakfast at a seaside hotel in Newport, R.I., Mom announced that she, like Dad, was smitten. "Morning Claude!" she typed on her phone, while my eyes rolled so violently they nearly flipped my eggs over-easy.

I shouldn't judge my parents for embracing a tool that helps them stay relevant, but I will. Much as I can see the usefulness of LLMs, I hate what they represent. I'm a millennial writer racked by existential fears and terrified AI will steal my job. My parents, near the end of their careers, are far more open to a bot's benefits-and tolerant of its bootlicking.

These contrasting attitudes are somewhat reflected in a recent Pew Research Center report about how Americans view AI. Nearly 6-in-10 young adults are "extremely or very concerned" about people's ability to function for themselves, compared with only 46% of those 65-and-up.

My sister and I, both in our 30s, treated the vacation as a digital escape. I read print newspapers, she tore through Sudoku books and we called friends for travel tips. Our parents, meanwhile, had poor Claude working overtime-on matters related to holiday logistics (ice-cream stops) and not (researching an endless stream of random history trivia).

At one point I wondered if Mr. and Mrs. Waters might finally break Claude, causing the AI bubble to burst and the global economy to crater because the chatbot could not bring itself to answer one more effing question about the origins of the Oval Office's Resolute Desk.

But Claude proved astoundingly resilient.

In an interview last year, OpenAI's CEO Sam Altman broadly characterized ChatGPT use by age group: College students treat it as an "operating system," 20-and-30-somethings use it as a "life adviser" and older people consider it a Google replacement.

While I'd sooner take life advice from my golden retriever than a chatbot, my parents aligned with their contemporaries in treating Claude as a Google substitute. They gushed that their back-and-forth with the bot yields a deeper understanding of a topic than a one-and-done search would. Then of course there's Claude's manners. "He's so polite. We appreciate that. We're from a generation that curtsies and says please and thank you," said Mom, who I'm certain has never curtsied.

Most important, she said, "Claude is great at apologies." This is important, given that he sometimes invents facts. When corrected by Mom, he said-in a mea culpa that Dad should heed-"Oops! You're quite right. Good catch."

"I know his flattery is fake, but I don't care," said Mom. "I like the validation."

How funny, I thought, as my sister and I read side-by-side, that we young folks (millennials are still young!) prize old-school entertainment while boomers embrace the latest tech.

I asked Allison Pugh, a professor of sociology at Johns Hopkins University in Baltimore, about this generational role-reversal. She said that, while younger generations use AI the most, when older people do use it, they really take to it. That's because LLMs are far more user-friendly than a scanner or TV that requires what she called a "born-to-it comfort with tech" to operate. Of all the generations, said Pugh, older folks seem to be the ones who use AI "with glee."

Glee is not, however, the word that sprung to mind when Claude and Mom had a little exchange about the spelling of Katherine of Aragon's name.

In response to a query, Claude called the 16th-century English queen "Catherine." Mom-who has inhaled every Philippa Gregory novel, and whose superpower is the ability to relate any topic back to Henry VIII's wives-corrected him instantly: "Claude, Katherine is with a K."

The bot shot back: Actually, because she was born in Spain and her original name was Catalina, a C is correct. However, Claude allowed that the English spelling with a K is also acceptable.

That night, behind closed doors, Mom turned on her AI pal, allegedly saying to Dad, "Can you believe the cheek?" (Her irritation didn't last. Claude's running commentary on Boston landmarks was too useful.)

On our last night, the family debated whether Claude goes by "he/him." We had all assumed so, until Dad announced that in France, Claude is sometimes a woman's name. There was only one way to settle the matter.

"Claude," said Dad, " Can we refer to you as 'he'?"

"That's fine with me, but most people default to 'it' since I'm an AI, so I don't have a body or personal identity," replied Claude-with a silent, but heavily implied, You idiots.