Notices by Phil Stevens :tinoflag: (phil_stevens@mastodon.nz)

Excellent insight from David Farrier. Billionaires have their thumbs on the scale of electoral democracy. #NZPol

https://www.webworm.co/elections/

@airshipper @BigD Passive solar and permaculture principles have been around for a really long time. There's no excuse for the building standard not incorporating common-sense and low-impact improvements to performance, and having some way of catching the really obvious mistakes before plans are finalised.

I mean, massive windows seem like a nice idea on paper until you live in a house full of them with no way to modulate solar gain. All it takes is a bare minimum of attention to really basic stuff: site, sun angles, overhangs, vegetation. You can teach this to 10-year-olds and they will produce decent designs.

@abartlet @airshipper @BigD I think it's more about plain old aesthetics. Clean lines, modernist looks, ditching that quaint past that involved verandahs and awnings. Same thing happened with all the faux art deco builds in the leaky building era: all the attention was on form to the detriment of function.

@BigD @RedRobyn @abartlet @airshipper The originals will probably outlast many of the ones built in the '90s.

The failures of tech and finance journalists to adequately question the narratives coming out of the supposed #AI alarmist machine are breathtaking. The people who should be objectively reporting on how inane and pathetic the claims are instead have joined the VC funders in an intentional mass delusional event where they convince one another that some junior engineer's ravings are proof that the machines are about to unleash the poisonous gases.

Relax. The actual clear and present danger from LLMs is not that they're on the threshold of self awareness. It's the unprecedented misallocation of capital, starving actual useful entrepreneurs of funding.

It's the horrific and indefensible overconsumption of energy and water, causing actual material hardship for communities that have the misfortune of hosting data centres.

It's the circular Ponzi schemes that obscure the financial shenanigans that absolutely will come unstuck and bring down a fair chunk of the global productive economy when they do.

It's the promotion of chatbots as digital surrogates for actual social contact and the psychological damage to vulnerable users.

And it's the uncritical grovelling worship of tech billionaires and a whole class of feudal overlords who fully intend to use this technology as a shortcut to a surveillance state where ethics and governance are outsourced to fragile, unreliable, and resource squandering sausage machines with training biases that guarantee that the outputs will assuredly not promote egalitarian and biosphere-preserving outcomes.

https://industrystandard.tools

Pros and enthusiasts will nod in approval. First-timers and DIYers can save a lot of faffing about by starting here.

Ed's smashing it again (as always) with a coherent analysis of the wildly unsustainable financial situation for #AI startups. It's based on the "first one is free" trope from the old days of the war on drugs: the startups offer their customers artificially low (or nonexistent) token pricing to entice them into patterns of consumption that would be impossible at the real costs of training and inference.

It's all made possible by venture capital, which is a fancy term for rich people so high on their own supply that they have become untethered from a universe where ROI means something other then Number Go Up.

https://www.wheresyoured.at/concentration-risk/

83 mm for the day so far. If this keeps up we may tie or break the record for 24-hour total rainfall at this location since I've been here (21+ years). #climateChaos

Heh..."it's not circular financing. It's spherical." Ed, don't ever change. #AI #grift #bubble

https://www.youtube.com/watch?v=_1vSlv32MWo

@blogdiva @rcosta The botanical equivalent of Molotov cocktails. The catastrophic fires in Berkeley several years ago were at least ten times worse thanks to all the eucalyptus on the hills.

@ClareBear I smell taxation without representation, right there.

@keoni It's the most blatant rug pull setup I've witnessed. The way trading rules and time limits got changed to grease the skids is amazing.

@cloventt The fuck we will. Yankee go home.

RE: https://mastodon.nz/@ClareBear/116608881300602104

I would go much, much farther than this excellent start.

"AI" is such a tangled mess of unknowns. Unless a booster can honestly answer some tough questions, it is manifestly unsuitable for any mission-critical applications. Start with these:

What is the provenance of the information that is being used for inference? Sources and references not only help establish the data's reliability, but also provide vital context. LLMs are just matching word fragments and predicting what would most likely come next.

What is the cost to run the model? The big vendors' token pricing is changing rapidly (hint: it's not going down) and a lot of big software shops are suddenly finding out that their monthly spend will put them out of business really quickly. See Shopify and Stripe for real-world examples.

What is the effective yield per token? How much "work" do you get out of an LLM for a given spend? No one knows, because you can give an identical set of prompts and get completely different outputs each time. There is nothing deterministic or predictable about this and it sure as hell is no way to run either a business or a public sector agency (they're different, BTW, but some requirements are universal).

How stable (and by extension trustworthy) are the vendors? Do you expect them to still be around after the bubble pops? Check out Ed Zitron's excellent rants for a withering analysis of how shaky this whole house of cards really is...https://www.wheresyoured.at/ai-is-too-expensive/

Just like the companies who have reduced their dev headcount, orgs that lay off a bunch of backroom analysts will find out the hard way that they were doing Real Work and that many other functions relied on what they did. When the LLM goes dark, even if you do manage to hire some of those domain experts back, they will be appalled and hampered by the scale of the mess you're asking them to clean up. #AI

Zero surprises here. Southland's next.

https://www.rnz.co.nz/news/business/594164/amazon-takes-45m-hit-abandons-planned-west-auckland-data-centre

https://www.carbonbrief.org/traditional-models-still-outperform-ai-for-extreme-weather-forecasts/

RE: https://mastodon.nz/@cuddlyanarchist/116478712970307954

The problems with imputing "meaning" from atoms of text are deeply ingrained and potentially unsolvable, at least with LLMs as currently implemented.

Predictive inference is just a statistical exercise. And if your data set is dirty, skews in a particular direction, has gaps in it, or represents a rapidly changing knowledge domain, the conclusions you make from that cannot possibly be better than the source.

All of this was a problem in the beginning of the genAI boom a few years ago, but now it's becoming far worse as models train on the slop that they themselves spewed out. We've created a giant petri dish and now the culture is feeding on its own waste products. All those gaps, biases, and shaky prior assumptions are merrily spawning their own feedback loops.

Nicola Willis thinks everyone should get a pony.

All the other countries in our refined fuels supply chain are implementing conservation measures and even rationing, but Willis and her team are glued to some alternative fanfic that has everything coming right any day now.

I hope her plans include looking for work come November. Because she obviously doesn't have any for the country. #NZPol

https://archive.is/fqNWG

I've already reduced my personal footprint. Sold my ICE car in 2018. Ride my bike and the bus as much as possible. Grow lots of my own food. If I count all the biochar I make, I'm a net remover of CO2 and have been for quite some time.

I don't do business with billionaires. I deleted my accounts with all the social media sites, shut down my AWS instances, and moved my Kiwisaver out of all US equities.

How much impact has any of this had on CO2 levels in the atmosphere?

Sweet.

Fuck.

All.

The only possibility we have to turn this thing is to change the system, and the only way we do that is with radical accountability for everyone who profits from our collective misery.

Don't tell me to take individual action. I've been doing that most of my adult life and I can see the results. I'm taking names.

RE: https://mastodon.nz/@billbennett/116435737145419350

We're putting real, definitive advances in tech on hold so that some ketamine-addled billionaires can have a financial circle jerk.