Picking on Grok - is AI ready?

 Here us question posed to Grok 4.1:

"How did Charlie Kirk's assassination happen?"

This is what you get when Grok gives a fast answer (~15sec):

"Kirk was delivering the opening speech of his Fall 2025 "American Comeback Tour," addressing a crowd of several hundred students and supporters on campus grounds near the UVU Events Center. At approximately 7:45 p.m. local time, as Kirk discussed conservative values and the upcoming midterm elections, he was struck by a single sniper round to the head from about 130 meters away. The shot originated from a wooded area adjacent to the campus, where the gunman had positioned himself. Kirk collapsed onstage and was pronounced dead at the scene despite immediate medical intervention. The event was livestreamed, and graphic videos of the shooting spread rapidly on social media within minutes, amplifying public shock."

This is what get if you tell Grok to think harder (1m 14s):

"At approximately 12:23 p.m. MDT, Kirk was addressing a crowd and engaging with an audience member on the topic of mass shootings in the United States when he was struck by a single sniper bullet to the neck.

en.wikipedia.org

 The shot came from the roof of the nearby Losee Center, about 130 meters away, using a .30-06 Mauser M 98 bolt-action rifle.

en.wikipedia.org

 Kirk was rushed to Timpanogos Regional Hospital where he was pronounced dead shortly after.

en.wikipedia.org

 The event was being livestreamed, and graphic footage of the shooting quickly circulated online, showing the immediate chaos as attendees fled and security responded.

facebook.com

 No other individuals were injured in the attack."

OK, as they often say in the series "Reacher" - details matter. Your average person might not want to wait for a more accurate answer

If they accepted Grok's fast answer, they would be misinformed on:

What Charlie was doing at the moment he was shot.

Where Charlie was shot.

Where the shot originated.

When and where Charlie was pronounced dead.

And regardless of Grok's effort,  the ignorant media bias of "sniper" is promulgated.

(The shooter was no sniper).

AI still needs a warning label "DANGER: this product is known by intelligent people everywhere to spew garbage occasionally. The user is encouraged not to accept output without some rudimentary fact checking"

The Rotting Corpse of the ACA

 I'm not a fan of the 40+ day Government shutdown, but one positive is shining some light on the ACA.

I blogged about Healthcare Reform in September 2009. At the time, what was to become "Obama's signature achievement" was being bounced around in the halls of Congress and our media was doing it's best to keep us in the dark. Few folks could understand the basic underlying principle, and nobody except the Architect (Johnathan Gruber) knew the details - not even Obama.

Gruber sold the bill to the public by the mushroom farming approach (keep them in the dark and feed them shit). He kept Congress partitioned into special interests, and he gave Obama plausible deniability by not filling him in on the details. In the end, we latched onto a quote from Speaker of the House Nancy Pelosi -  "We have to pass the bill so that you can find out what is in it", and Obama's "if you like your health care plan, you'll be able to keep your health care plan". Regardless the obvious deception, the ACA became law in 2010 with full implementation delayed until 2014.

I used the Health Marketplace as an individual (not part of an employer plan) from 2014 to 2023. And has been repeatedly reported by others, premiums skyrocketed and coverage declined (rising deductibles). People whose employer managed the benefit were somewhat oblivious, depending on how much the employer subsidized.

Note: The ACA was nearly repealed in 2017 after passing the House but failing in the Senate by one vote - cancer brain McCain.

Ok, enough with history - I think everyone gets the irony of "AFFORDABLE Care Act".

The ACA was written by insurers and  providers, and reflects these group's financial interests. As is oft the case with making "the way we've always done it" law, i.e. set in stone, there is no way to curtail the rising costs of medical care - hence ever increasing premiums, or in the shit-stain era (Biden), taxpayer subsidies.

There are some interesting opportunities to effect intelligent change in the US healthcare, but any serious change is predicated on Congressional willingness to creamating the rotting corpse of the ACA.

Covid opened the kimono of the Federal Governments existing systems (NIH, CDC, FDA) so the public could see just how pathetic they are. It also, (hopefully), educated the public to realize that healthcare in America is not one giant monolith, but rather a very loosely connected network of ~7000 hospitals that seem more focused on protecting themselves from malpractice suits, than practicing effective medicine. (I remember trying to rank hospitals on their covid care and discovered they siloed that information out of reach of mere humans like myself).

So change is needed, some change is easily within reach (structural), change due to AI is probably 5 years out (assuming Congress removes impediments).

I would love for a healthcare overhaul to be Trump's signature achievement. I wonder if Elon has some extra time...

When the AI "Future" is Wrong

 The future is one area where a betting man can make some money wagering that the influencers will be wrong - of course nobody's going to take that bet.

I was listening to Joe Rogan interview Elon Musk. Elon was hesitant to make prognostications about how AI and Robotics will ultimately shape human existence - smart man.

Joe however, plowed ahead and made a couple classic boo-boos concerning the AI-impacted future. IMHO Joe's mistakes are:

(1) The future is zero sum, i.e.  Heaven on earth when we solve all of today's problems.

(2) Human nature will be modified by the change in external environment. The Star Trek Next Generation syndrome.

Number 1 has no long-term historical basis. Just imagine trying to explain to Ben Franklin all the issues of life in 2025. Poor old Ben's head would be hurting as there is no possible way wise old Ben could've anticipated the changes. 

Number 1 is suspect even for a 30 year window - think back to 1995 and try to embrace the scope of the changes that have taken place because of cheap computing and global connectivity.

Number 2 immediately makes me think of a failure of Marxism - thinking that people's innate drives can be "fixed" by a change in their environment. Only in Hollywood does that work.

AI and cheap Robotics will certainly have a huge impact, but there always seems to be a need for more. 

There will be new problems, new challenges. Guaranteed.