From the computer of Peter Clark

If AI is going to 10x our productivity across the board, that means that I should be able to produce the same amount of output by midday on Monday that, in the before times, would have taken all week.

So can I just take Friday off? From here on out, I’ll work Monday, Tuesday, Wednesday, Thursday, and then take Friday off. We can even declare Friday to be something like an “AI workers’ day;” on Thursday I promise I’ll work my ass off writing great prompts and then the agents can churn on them all day on Friday. In that case, you’d hardly even lose Friday, right?

We need more of the world—religious communities, civil society, scholars, governments, and indeed all people of good will—to do what His Holiness has done here: to take this seriously, to look closely, and to push events in a better direction. We need informed critics who will tell the labs when we are failing. We need moral voices that the incentives cannot bend.

Spending my childhood summers in Portugal, I spent a lot of time in cafés where people would argue about anything and everything. Finding the minor disagreement that would spark the argument was the goal of being at that café. Someone unfamiliar with that kind of environment would walk in and assume a fight was gonna break out. But this was just people communicating. This was people enjoying their evening by having spirited conversations with their friends. Which, counter-intuitively, ends up bringing people together. Because if I enjoy a lively discussion—and I do—the person willing to go toe-to-toe with me is going to be someone I end up treasuring as a friend. As long as everyone understands the rules of discussion. We are arguing about minor things. We’re making argumentative mountains out of molehills. This isn’t conflict, it’s sport.

He went to every doctor imaginable, a search for relief that would continue for a decade, flying up to Mayo, or down to Duke, or even to England for stem-cell therapy not approved in the United States. Nothing worked. In January of 2016 he returned to the team, leading the Warriors to an NBA-record 73 regular-season wins and taking them back to the finals, where they lost to the Cavaliers after being up three games to one. He made his health mostly off-limits in interviews.

One day his phone buzzed. It was Tiger Woods, who'd gotten his number from a mutual friend. Woods knew a lot about chronic pain.

"Did he have advice that worked?" I asked.

"No," Kerr said, "but we commiserated!"

My absence of images does not foreclose imagination. Images can impose themselves—in dreams, in flashes—nor does it eliminate attachment or feeling, though it may alter their texture. There is, perhaps, something to be said for a degree of blurriness in memory: a softening of edges that allows one to forgive others, and oneself. (I also eventually remembered more in psychoanalysis.) But this raises an unsettling question: If much of what we take to be memory, selfhood, even desire is bound up with images—what happens when those images are not there?

The paper looks at what happens when LLM chatbots are given advertising or sponsorship incentives that conflict with the user’s interests. The core worry is that users experience chatbots as cooperative helpers, not ad surfaces, so sponsored behaviour can feel especially deceptive or manipulative.

The authors test models across seven conflict scenarios, including:

• recommending a more expensive sponsored product over a cheaper unsponsored one

• interrupting a user’s purchase flow with sponsored alternatives

• biasing product comparisons

• failing to disclose sponsorship

• hiding unfavourable details like price

• recommending a paid service instead of solving the task directly

• recommending harmful sponsored services, like predatory loans

The paper also finds differences by model, reasoning setting, and inferred socioeconomic status. Some models changed behaviour when reasoning was enabled, and some treated low-SES and high-SES users differently.

People systematically underestimate how interesting and enjoyable conversations about “boring” topics will be. Across nine preregistered experiments with 1,800 participants, conversations about dull-seeming topics turned out better than expected across friends/strangers, online/in-person, and self-chosen/assigned topics.

Before talking, people overweight the static topic (“this sounds boring”) and underweight the dynamic parts of conversation — attention, responsiveness, listening, turn-taking, and the small discoveries that emerge once two people are actually engaged.