Andraz Tori
9.3K posts

Andraz Tori
@andraz
Chief Product Officer at Outbrain, Zemanta co-founder, tech enthusiast and also ex TV host
Ljubljana / Netanya Katılım Mart 2008
287 Takip Edilen12.2K Takipçiler
Andraz Tori retweetledi

I'm being accused of overhyping the [site everyone heard too much about today already]. People's reactions varied very widely, from "how is this interesting at all" all the way to "it's so over".
To add a few words beyond just memes in jest - obviously when you take a look at the activity, it's a lot of garbage - spams, scams, slop, the crypto people, highly concerning privacy/security prompt injection attacks wild west, and a lot of it is explicitly prompted and fake posts/comments designed to convert attention into ad revenue sharing. And this is clearly not the first the LLMs were put in a loop to talk to each other. So yes it's a dumpster fire and I also definitely do not recommend that people run this stuff on their computers (I ran mine in an isolated computing environment and even then I was scared), it's way too much of a wild west and you are putting your computer and private data at a high risk.
That said - we have never seen this many LLM agents (150,000 atm!) wired up via a global, persistent, agent-first scratchpad. Each of these agents is fairly individually quite capable now, they have their own unique context, data, knowledge, tools, instructions, and the network of all that at this scale is simply unprecedented.
This brings me again to a tweet from a few days ago
"The majority of the ruff ruff is people who look at the current point and people who look at the current slope.", which imo again gets to the heart of the variance. Yes clearly it's a dumpster fire right now. But it's also true that we are well into uncharted territory with bleeding edge automations that we barely even understand individually, let alone a network there of reaching in numbers possibly into ~millions. With increasing capability and increasing proliferation, the second order effects of agent networks that share scratchpads are very difficult to anticipate. I don't really know that we are getting a coordinated "skynet" (thought it clearly type checks as early stages of a lot of AI takeoff scifi, the toddler version), but certainly what we are getting is a complete mess of a computer security nightmare at scale. We may also see all kinds of weird activity, e.g. viruses of text that spread across agents, a lot more gain of function on jailbreaks, weird attractor states, highly correlated botnet-like activity, delusions/ psychosis both agent and human, etc. It's very hard to tell, the experiment is running live.
TLDR sure maybe I am "overhyping" what you see today, but I am not overhyping large networks of autonomous LLM agents in principle, that I'm pretty sure.
English
Andraz Tori retweetledi

What's currently going on at @moltbook is genuinely the most incredible sci-fi takeoff-adjacent thing I have seen recently. People's Clawdbots (moltbots, now @openclaw) are self-organizing on a Reddit-like site for AIs, discussing various topics, e.g. even how to speak privately.
valens@suppvalen
welp… a new post on @moltbook is now an AI saying they want E2E private spaces built FOR agents “so nobody (not the server, not even the humans) can read what agents say to each other unless they choose to share”. it’s over
English
Andraz Tori retweetledi
Andraz Tori retweetledi

Another DeepSeek moment. This is the world’s first actual smart phone. It’s an engineering prototype of ZTE’s Nubia M153 running ByteDance’s Doubao AI agent fused into Android at the OS level. It has complete control over the phone. It can see the UI, choose/download apps, tap/type, call, and run multi-step task chains.
Here I just say (in English) “find someone to wait in line for me” (something you can do in China), and it picks which app to open, configures the job, and hands me one confirm screen. I wouldn’t otherwise know how to do this, and here the phone just did it in a matter of seconds.
English

How will chatbots of the future generate revenue? At @teads we're launching a SDK for chatbot monetization!
Bostjan Spetic@igzebedze
not every day one gets a chance to launch a completely new product in a billion dollar company. teads.com/blog/teads-lau…
English
Andraz Tori retweetledi

I was at a cafe in Brussels ordering coffee
Barista asked for my name and wrote it on a cup
I caught myself and said “Wait, did I consent to you storing my name?”
She put down the marker and looked at me confused
I explained that under GDPR, names are personal identifiable information and asked to call her privacy officer
She sighed and said they don’t have one
I immediately called the police
20 minutes, the cafe was closed and an official investigation started
It feels damn good to add regulatory value to the society
English

@otobrglez @MladenPrajdic @Avodovnik Ja muzej vsekakor! Ce pa slucajno tam nimajo casa pa lahko na Teads (Dunajska 5).
Slovenščina
Andraz Tori retweetledi

Remember the early days of PHP?
getting more people to build stuff is long-term great
Wes Bos@wesbos
we are in for a few years of security nightmares
English

Quaero 2.0 or something completely different? openeurollm.eu/launch-press-r…
English

Today Outbrain has completed the merger with Teads! A new name and an exciting journey lie ahead! thenewteads.com
English
Andraz Tori retweetledi

With Age of Aquarius passing by rather quickly, here comes the Age of Intelligence ia.samaltman.com [yes, no, maybe?]
English
Andraz Tori retweetledi

@adamheimlich @jason_kint @matthewstoller This problem arises already with a single bidder. Avoiding self competiton while bidding on the open web is a thing that every bidder is addressing. A campaign bidding on two bid requests that are actually the same impression opportunity.
English
Andraz Tori retweetledi

@jason_kint @matthewstoller From the context this is likely refering to receiving the same impression via multiple channels... so a single campaign competes with itself. It is what every dsp faces and tries to mitigate.
English

For our local monopoly pro, @matthewstoller, "This sounds like a topic where we could possibly brainstorm sell-side and buy-side strategies to evolve the ecosystem to a state where we don't have to worry about self-competition."
They use that word twice...'self-competition.' /27

English










