Matouš Hrdina

Newsletter editor at @SeznamZpravy. Avid Wikipedia reader. Newsletter Pod čarou:

[Filip Harzer @HarzerF]

Polské ministerstvo vnitra právě zveřejnilo dvanáctistránkový dokument, který popisuje, jak reagovat v případě útoku ze vzduchu. Zmiňuje například zásadu dvou stěn. Instrukce jsou dostupné online a dostanou je i všechny obce. https://t.co/Pj8PvFYuct

[Iuridicum Remedium @iure_cz]

Zatímco síť Traficon využívá skenování obličejů od letošního léta k cílení reklamy, na festivalu Beats for Love se tváře návštěvníků nově rozpoznávají, aby se zamezilo podvodům s lístky. Podle naší analýzy může jít o porušení zákona.

👉 https://t.co/I2arLa6CWo

[Matouš Hrdina @MatousHrdina]

https://t.co/kirsbbUSVD

[Matouš Hrdina @MatousHrdina]

V debatách o údajných rizicích zdivočelé AI schází nejen v Česku víc ekonomického pohledu - třeba fakt, že Anthropic a OpenAI se skrze vyvolávání paniky snaží prolobbovat kartelovou dohodu a ochranu před čínskou konkurencí. Víc v novém newsletteru👇

[Laurel @BalanceCrafting]

Jonathan Kanter on AI

[kate wehwalt @mcmansionhell]

extremely fascinated by the fact that the 'soulless' architecture in these kinds of videos is increasingly no longer communist-era blocks but post 1990 world bank neoliberal dreck

[@timnitGebru (@dair-community.social/bsky.social) @timnitGebru]

Welcome all the security people coming this way. Make yourselves comfortable because they’ve been doing this bullshit since the beginning. They coined the term “AI safety” to differentiate themselves from those of us conducting empirical research & pontificate about magical beings

[zeynep tufekci @zeynep]

And y’all laughed when people pointed out that generative AI was NOT going to make Big Consulting go bankrupt, not because LLMs can’t produce very similar reports but because the key function of many such companies is providing legitimacy and a social stamp to existing choices.

[Taylor Lorenz @TaylorLorenz]

The chances of catastrophic harm from AI are so much more likely to come from government misuse/fuck up than some wild theoretical sci fi x risk scenario

[Nate Wolff @proffwolff]

Stop turning structural problems into men's personal failings! This was the enitrely predictable result of the decline of the whale fishery.

[The Washington Post @washingtonpost]

Brent Hecht, Microsoft’s director of applied science, called the training of AI models the “largest theft of labor in human history,” according to a legal brief. https://t.co/gsCCwFOgFc

[Megan Gray @megangrA]

and deposition transcript from Microsoft CEO Satya Nadella that AI hover of paywalls news is copyright infringement (requires license). Lordy

https://t.co/I5D4CTQ1qN

[michal špína @michalspina]

chystaná výměna vlaků @ceskedrahy_ za @RegioJet na trati Praha – Hradec – Trutnov ukazuje, jak se model soutěží na železnici vyčerpává. všichni tak nějak vědí, že RJ bude horší než ČD, málokdo ho tam chce, ale prostě nabídl státu větší lowcost a my cestující s tím nic nenaděláme.

[Ken MacLeod @amendlocke]

The Rouse by China Miéville review – a mighty achievement from the ‘new weird’ author | Fiction | The Guardian https://t.co/kv5leLQtWR

[Arnaud Bertrand @RnaudBertrand]

This very much sounds like a West-only AI cartel proposal and, in fact, if you read Dario's essay - under the very telling "pacing within democracies" chapter - he specifically says he'd need the US government to issue an "antitrust waiver" for this.

The plan transparently looks like this: agree among themselves not to compete too hard, get an antitrust waiver to make it legal, and while they're at it, get Washington - under the veneer of "safety" - to kneecap Chinese competitors because only "democratic" models can be in this "safe pacing" club.

I'm not even exaggerating: Dario pretty much writes this explicitly.

There are 3 steps to his plan and step 2 is literally: "Frontier AI companies within democratic countries coordinate to establish common safety standards as well as limits on the rate of unchecked AI progress" -> the "unchecked AI progress" that would need "limits" is simply AI progress that happens outside their agreement. Another word for it is simply competition 🤷‍♂️

He also explicitly writes that "a key part of pacing within democracies is to keep democracies’ AI lead over autocracies as large as possible." You can't possibly be clearer than this...

[Alonso Gurmendi @Alonso_GD]

You’re seriously telling me you believe AI workers with access to super secret dangerous tech have no NDA in their contract and the reaction to them accusing their former employer of risking the future of humanity is the CEO saying “yeah we should slow down”? Bc they “care”? Lmao

[michal špína @michalspina]

„no jo, o tom už jsme psali,“ říkáváme často v A2.

takže pokud se teď ptáte, co všechno vedlo k volebnímu úspěchu AfD, doporučuju číslo @ZkrA2tka Východní Německo, které jsme vydali na podzim 2019 k výročí pádu Berlínské zdi. je kompletně odemčené na našem webu (link níže).

[Vojtech Pecka @VojtechPecka]

Dnes je v Číně elektrický už 40% z nově prodaných těžkých náklaďáků.

V ČR zatím řešíme, jak si co nejdéle udržet spalováky.
Oproti Číňanům budeme moci déle:
– dýchat toxické zplodiny
- být obtěžováni hlukem
- být závislí na fosilním importu z nepřátelských zemí.

Hurá.

[Matthew Green @matthew_d_green]

This was always going to happen to age verification systems. It wasn’t even so much of a prediction as it was an inevitability. The second half of this is tying actual web usage to real-world ID, and breaches will give us that too, soon.