The Honest Courtesan

We need to take back control over the tools that we use.  –  Mike Masnick

It’s a Start

Let’s see if this lasts any longer in Pittsburgh than it did in New Orleans:

Pittsburgh City Council [has passed] an ordinance that would [allow cops to choose to charge]…sex work[ers with]…a [municipal] summary offense [instead of the state misdemeanor charge, which]…“can involve up to a year in jail, and…fines [of]…up to $10,000”…the m[unicipal offense would inflict]…“just a small fine similar to a traffic ticket”…[state] laws also require those charged with prostitution to show up to court, a requirement that the [optional offense under the] ordinance would do away with… the ordinance would not [remove cops’]…ability…to…charge people [with the existing state prostitution offense]…

These halfway measures are just a bone to activists; they never make a difference and never last more than a couple of years.

The Puritan Recrudescence (#1410)

Burble burble BLUE STATE burble drool:

If [politician]s in New York and Minnesota get their way, social media users will be [harassed by compelled-speech popups making nonsensical claim]s about the [supposed] toll their scrolling habits could be taking on their mental health.  Legislatures in both states recently passed legislation requiring [“]warning labels[“] on social media, despite a…federal court…declaring a similar law for porn websites…unconstitutional.  The Minnesota [version] says a “conspicuous mental health warning label” must appear each time a user accesses a social media platform and only disappear when the user exits the platform or [clicks a box]…The New York [version]…mandates that any social media platform that provides [typical] basic features…must display a warning label “prescribed by the commissioner of mental hygiene.”  This is peak performative lawmaking [which]…lets [politicians]…pretend to be doing something to save us from the boogeyman du jour, Big Tech…In the real world, government goons telling you to put down Instagram and go touch grass would be unlikely to help even if it could pass constitutional muster—which it almost certainly won’t…

The Cop Myth (#1546)

Why are people shocked when men paid and encouraged to behave violently, behave violently?

A [North Carolina cop] resigned [after] he threatened a man with [a knife] at a restaurant…and accused [his victim] of touching his daughter.  Wesley Lane[‘s victim] has epilepsy, autism and an intellectual learning disability…surveillance video and witness statements [proved Lane’s excuse was full of shit]…

You Were Warned (#1550)

The subtitle above says it all:

…Starting at the end of 2025, Australians [who don’t use VPNs] will [be forced] to verify their ages when…using e[ither]…Google [or] Microsoft [Bing, which were singled out as] the two biggest search engines in the country…search engines [must otherwise “impose] tools and/or settings like ‘safe search’ functionality, at the highest [censorship] setting by default[, assuming every user]…is…an Australian child”…the measures are similar to what’s being considered for the country’s under-16 social media ban…age assurance laws like this…[are known to] be [in]effective[, but politicians don’t care because the “internet safety” theater] is [the entire point]…VPNs allow people to surf from other locations, bypassing location-specific rules…and [sites not based in Australia] may simply choose not to [infantilize their users]…

Mad Libs (#1550)

These algorithms are not only unintelligent; they are making those who use them look stupid as well:

…there’s now something of a cottage industry for writers and coders who specialize in fixing [LLM vomit]…Sarah Skidd, an American product marketing manager…was approached by an agency that urgently needed someone to redo copy…after having a…chatbot do the work to save a few bucks…she spent 20 hours redoing the copy from scratch…[which] meant her client was shelling out $2,000 for copy that likely would have ended up being far cheaper had a human just written it in the first place.  With more and more companies rushing to [jump on the fake “]AI[” bandwagon] over the past few years, things are shaking out in [100% predictable] ways…Sophie Warner…of the UK-based digital marketing agency Create Designs..[ha]s been fielding more and more requests from clients to clean up [LLM vomit]…In one recent case, a client ended up going without a website for three days and having to pay Create Designs nearly $500 over a small line of code that ChatGPT had written for them…[when a human could have written] the original…[in] about 15 minutes…but [gullible fools are still buying tech advertising claims]…that [computer programs] can replace expert work…

Torture Chamber (#1551)

This will continue until the evil US immigration policy is reformed:

Immigrants being [locked up] in [large closet]s in Lower Manhattan have [reported] being unable to bathe or change clothes, cramped conditions, sometimes being provided just one meal a day, and sleeping on concrete benches or the floor…[for] days at a time… “There’s no room to sit down – standing room only,” said Rebecca Rubin, an immigration attorney…[cowardly] Congress[critter]s, who for weeks have been refused entry at the site…have [merely complained rather than using their authority to enter and inspect]…The [closet]s used to be temporary holding areas where [victims of ICE] were [locked] for a few hours before being transferred to larger…[prisons]…But…in recent months, [victims] have been [packed in like sardines]…for [up to 10] days…

Aladdin’s Satellite

How to make an LLM simulate psychosis:

[Last] week, Elon Musk’s Grok [chatbot] started spewing extreme antisemitism, responding with conspiracy theories about Jewish people, and for a brief period telling people to call it “MechaHitler”.  The incident perfectly illustrates why Alex Komoroske’s manifesto about the dangers of centralized AI, which we ran less than a month ago, has been making waves.  When a single person controls the dials on an [LLM] system, they…inevitably will…tweak those dials to serve their own interests and worldview, not their users’.  Just days [before], Elon claimed that his team had “improved Grok significantly” and that “you should notice a difference when you ask Grok questions”…People sure did notice a difference.  The transformation wasn’t subtle, and it wasn’t accidental…a similar [ham-handed “improvement”] two months or so ago [resulted in] Grok bec[oming] obsessed with linking everything to [“]white genocide[“]…as the situation escalated, they [tried to fix it, but]…eventually…[had to take it] offline entirely…

From now on, house style on The Honest Courtesan will be to refer to Musk’s chatbot not by the wildly-incongruous name “Grok”, but by its self-chosen appellation “MechaHitler”.

I find paywalls distasteful, and so many people find this blog valuable as a resource I just can’t bring myself to install one.  Furthermore, I find ad delivery services (whose content I have no say over) even more distasteful.  But as I’m now semi-retired from sex work, I can’t self-sponsor this blog by myself any longer.  So if you value my writing enough that you would pay to see it if it were paywalled, please consider subscribing; there are four different levels to fit all budgets.  Or if that doesn’t work for you, please consider showing your generosity with a one-time donation; you can Paypal to maggiemcneill@earthlink.net or else email me at the same address to make other arrangements.  Thanks so much!

Read Full Post »