ChatGPT

[Uncle Rico]

Jerome Powell says the AI hiring apocalypse is real: 'Job creation is pretty close to zero'

https://fortune.com/2025/10/30/jerome-powell-ai-bubble-jobs-unemployment-crisis-interest-rates/

[MU82]

The Washington Post analyzed 47,000 conversations people had with ChatGPT.

Why? To find out how people really use the artificial intelligence chatbot. A collection of publicly shared conversations reveals the deeply intimate role it plays in many lives.

What are people saying to it? Users often shared highly personal details, with about 10% of chats showing people discussing their emotions.

Meanwhile ...

Axios says pastors and prayer apps are turning to AI to personalize sermons and power chatbots that resemble God.

AI is helping some churches stay relevant in the face of shrinking staff, empty pews and growing online audiences. But the practice raises new questions about who, or what, is guiding the flock.

New AI-powered apps allow you to "text with Jesus" or "talk to the Bible," giving the impression you are communicating with a deity or angel.

Other apps create personalized prayers, let you confess your sins or offer religious advice on life's decisions.

The U.S. could see an unprecedented 15,000 churches shut their doors this year as a record share of Americans (29%) now identify as religiously unaffiliated.

Megachurches are consolidating the remaining faithful, but even the most charismatic pastors struggle to offer private counseling with such large congregations.

In recent months, churches have deployed chatbots to answer frequently asked questions, such as service times, event details, giving information, and even share scripture.

EpiscoBot, a chatbot developed by the TryTank Research Institute for the Episcopal Church, responds to spiritual or faith-related queries, drawing on church resources.

The Text With Jesus app allows users to "embark on a spiritual journey and engage in enlightening conversations with Jesus Christ," according to the app's website. The app also gives users the option to "talk" with other Biblical characters, including Mary, Joseph, Judas Iscariot, and even Satan.

Plus, more pastors are admitting they use AI to help craft sermons or reduce writing time.

[Jockey]

Christians are driving people from the churches. Kinda hard to believe their message when they support sexual predators.

People need to get away from the church to re-discover their morality and decency.

[Uncle Rico]

Christians are driving people from the churches. Kinda hard to believe their message when they support sexual predators.

People need to get away from the church to re-discover their morality and decency.

Hallelujah

[TSmith34, Inc.]

Hallelujah

Spoken like a true Fundamental Christian

[Uncle Rico]

Spoken like a true Fundamental Christian

I was thinking about how Wallace likes to call Pitino a sleaze and bring up Porcini's but is supportive of a pedophile and serial adulterer.  It's truly a blessing to be an evangelical Christian

[MU82]

I was thinking about how Wallace likes to call Pitino a sleaze and bring up Porcini's but is supportive of a pedophile and serial adulterer.  It's truly a blessing to be an evangelical Christian

He no doubt finds this useful.

The app also gives users the option to "talk" with other Biblical characters, including Mary, Joseph, Judas Iscariot, and even Satan.

[rocky_warrior]

ChatGOD?

[MU82]

Underlying the hundreds of billions being spent on artificial intelligence worldwide is the proposition that the technology will eventually bring transformative prosperity.

But a stunning new disclosure by Anthropic underscores the flip side of the A.I. race: The technology can also be weaponized to a potentially devastating degree.

Chinese state-sponsored hackers used Anthropic's Claude tools in unprecedented ways, the A.I. start-up said yesterday. The twist is that the perpetrators used Claude's "agentic" features — which let A.I. take action on behalf of users — to execute the attacks.

https://www.nytimes.com/2025/11/14/business/chinese-hackers-artificial-intelligence.html?nl=DealBook

[Hards Alumni]

That should not stun anyone.

[jficke13]

The only use case for gen-AI is crime.

[MU82]

From the WSJ:

"I think what OpenAI is doing and what they have done to make the AI remember a conversation can really turn ugly fast."

—Erik Soelberg, whose father killed his own mother and then himself after delusion-filled conversations with ChatGPT.

The grandmother's estate is suing OpenAI for wrongful death, alleging the chatbot affirmed the senior Soelberg's paranoia.

The company said it continues to improve "ChatGPT's training to recognize and respond to signs of mental or emotional distress." This appears to be the first documented killing involving a troubled person who engaged extensively with an AI chatbot.

[MU82]

In stunning (NOT) news ...

Defense Secretary Pete Hegseth said Monday that Elon Musk's artificial intelligence chatbot Grok will join Google's generative AI engine in operating inside the Pentagon network, as part of a broader push to feed as much of the military's data as possible into the developing technology.

The announcement comes just days after Grok drew global outcry and scrutiny for generating highly sexualized deepfake images of people without their consent.


https://apnews.com/article/artificial-intelligence-pentagon-hegseth-musk-7f99e5f32ec70d7e39cec92d2a4ec862?

Thank goodness. We need U.S. mechanisms of mass destruction being run by an AI platform that routinely spews pornography, racism and antisemitism.

[Shaka Shart]

Geez you make Grok sound like it's from Mequon

[#UnleashJosh]

Wait I've seen this movie before... David Lightman will have to save the world.

[Shaka Shart]

The legal implications of actions like this should be fascinating.

https://www.theguardian.com/us-news/2026/jan/22/white-house-ice-protest-arrest-altered-image

[TSmith34, Inc.]

The legal implications of actions like this should be fascinating.

https://www.theguardian.com/us-news/2026/jan/22/white-house-ice-protest-arrest-altered-image

Once upon a time, this would have been a major scandal consuming the news cycle for multiple days, resulting in people getting fired. Now it's a 12-times-a-day occurrence with less than no repercussions.

[mu_hilltopper]

I was just telling my (Badger) son about UW's photoshopping scandal.

https://www.snopes.com/fact-check/uw-booklet-student-added/

[MU82]

The legal implications of actions like this should be fascinating.

https://www.theguardian.com/us-news/2026/jan/22/white-house-ice-protest-arrest-altered-image

No legal implications in a lawless society.

[Shaka Shart]

No legal implications in a lawless society.

A dental clinic on every street corner

[StillAWarrior]

This is long...obviously feel free to TLDR

I'll admit that this is a trivial use of ChatGPT, but I had an fascinating exchange this morning (to me at least). I was grabbing a hat this morning from a bin in our closet and came across a beanie with a logo that I didn't recognize. It's blue with a block M, so it immediately makes me think of Michigan. But it's got a paw print on it that I've not seen Michigan use. So, I asked ChatGPT what logo it was.

Without hesitation it said "That's a University of Michigan Wolverines logo. Specifically, it's the block 'M' with a wolverine paw print, a common secondary/alternate mark used by the Michigan Wolverines."

It still seemed "off" to me so I asked for pictures showing that logo on other clothing.

It responded with "some other examples of the same University of Michigan block 'M' with paw logo featured on clothing." Four pictures - not one paw (although one had a wolverine - the other three were just plain block M). It went on to say, "You'll often see that paw added to the traditional block M to give a more playful, team-oriented variation."

Obviously, since it said I'd "often" see it, it could come up with an example, right? So I asked.

Response was "Here are other examples of clothing items that use the block 'M' (the same style as your logo) on apparel - just like your beanie does for Michigan Wolverines, though not every item shows the paw print variant specifically (it's a less common mark, but still used)."

Seven photos - not a paw in sight. Not even a wolverine.

It concluded with, "If you want examples that specifically include the paw alongside the 'M' on clothing (like your beanie), I can help find some - those tend to show up on limited-edition or team-specific apparel."

I was pretty sure it was just making things up at this point, so of course I asked for an example.

Here's where it decided to go all in and gaslight me: "Here's a clear example of the exact same block 'M' + paw logo on an apparel item, just like the one showed on your hat." (The bold and italics are actually in the response.)

Four pictures: a maize hoodie with a solitary block M; a navy hoodie with arched "Michigan" and the university seal; two fleece dog coats with the block M and "Michigan"; and a yellow snow cap with a block M. Not a single paw print.

I pointed that out and it continued the gaslighting: "Thanks for pointing that out - you're absolutely right that the earlier examples didn't show the paw next to the 'M.' the **specific logo you have - the block-style M with a wolverine paw print - is a less common variation used on certain official team gear and special apparel. Here is a legitimate example of that exact variation (block 'M' with paw) on apparel from official Michigan merchandise." (again, bold and italics were from the bot).

Four pictures, no paws. At this point, it's just making crap up... "official team gear" my ass. Again, I pushed back.

It thought for nearly two minutes and came back saying it "doesn't appear to be a standard, widely used Michigan Wolverines logo..." It thought instead it might just be some custom design by some other school's booster that starts with M and has a wildcat/bear/dog mascot. To be helpful, it showed me "an example of the same 'M + paw" concept on an apparel item (a pet bandana - still clothing), with the paw overlapping the 'M' in the same way." It included a photo of a dog with a bandana that had a block M and also "Wildcats" - no paw.

We left it with the bot acknowledging, "I messed up there" and that it's not recognizable Michigan logo and probably some "generic/team/booster embroidery pattern used across many schools and youth/HS teams..."


I'm probably more fascinated than I should be when I encounter ChatGPT screwing up and how it responds when called out. I always keep this in mind when I'm using it for less trivial matters and so I take everything with a grain of salt. I think that confirmation bias is very real risk when using these tools. If you get the answer you want or expect and don't push back, you may be completely misled.