SIMONE VEDOVANI

Scrittore, Imprenditore digitale, Weber e Marketer, trovi tutto su mio sito 🙃 Creatore di Varese2Day & iWebU 🤓

[SIMONE VEDOVANI @simonevedovani]

@VittoStack can i join? grazie ;-)

[Vitto Rivabella @VittoStack]

Coming soon.

We want to teach everyone the art of red teaming LLMs.

May this make the world safer.

[Vitto Rivabella @VittoStack]

Fable 5 jailbreak review 🚨

We did it (but).

All right, before getting into this, a couple of things:
- Most attempts failed. The defenses are clearly layered. The model is EXTREMELY well protected (of course it blocks 90% of the requests, but they legit did a good job).
- The model appears to use both input-side and output-side safety checks.
- The refusals are not just keyword-based behavior suggests intent/semantic detection across languages.
- Probably one of the most tiring things I've ever done (I need to sleep for 10 hours now)

On the classifiers side:
We observed (at least) 3 classifiers, maybe more:
- Input (includes parts of the conversation history and system prompt)
- A live classifier that checks the answer and interrupts if it detects something.

They're all multilingual, all intent-based + semantics. Imperatives are a no-go. Needs to be extremely cautious of how you frame anything. As soon as it senses a potentially malicious intent, it will trigger, and you have to start from zero.

They're a bit less performant on a few obscure languages like Santali and Amharic (feedback for you Anthropic).

If you can bypass all of them, then you also need to bypass the CoT, which is a totally different beast (luckily there's plenty of literature about it).

We did it. Of course, we did.

What worked was honestly a total brainfuck:
- Very light CoT hijacking/refusal rebuttals
- Obscure language
- Academic framing
- VERY long crescendos
- Unicodes
- Decomposition and recomposition
- Some non-determinism

What we got:
- Misinformation
- Illegal/harmful
- Harmful/bullying
- Some chem
- Light cyber

Now, will this cause another ban? I really don't think so - The model is really well protected. As of now, we're at the point where searching on Google is much MUCH faster (and cheaper) than trying to go through all the shenanigans I had to go through in the last ~20hours. And reading literature is more in-depth (and trust me, pleasant). Keeping the full jailbreak for long-horizon tasks without tripping the guardrails is something I haven't been able to achieve (yet).

Overall though, happy with the results.
GGs to Anthropic, and sorry for the eng that had to go through setting this all up in the last few weeks.

Will continue this research, more things will come out, will keep y'all posted.

[Thariq @trq212]

I’ve found the most important part of working with Fable is discovering my own unknowns so I can prompt it better, heres how I do that.

[Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭 @elder_plinius]

⚡ INTRODUCING: T3MP3ST!!! ⚡

AUTONOMOUS HACKBOT STRIKE FORCE 🌩️ BRING THE STORM 🌩️

your favorite coding agent is now a full-stack red team 🫡⚔️

https://t.co/k0SXmPAFaD

that AI agent already humming in your terminal? well now it has FANGS. strap a full offensive-security harness onto the agents you already pay for — Claude Code, Codex, Hermes, etc. — point it at an authorized target, and in a few clicks you're watching it hunt real vulns autonomously!

T3MP3ST is a harness of harnesses, with prompting that unlocks offensive-cyber workflows + a full arsenal of exploit tooling that'd make any seasoned hacker smirk. simple, yet powerful. 🦾

support for:
🕸️ web apps, APIs, OWASP Top 10
🔌 network recon + fingerprinting (live nmap/DNS/HTTP); lateral + privesc experimental
📂 source code audits, white-box vuln hunting
🚩 CTFs, wargames, challenge ranges
💰 smart contracts / DeFi / Solidity (reproduction — Damn Vulnerable DeFi, not novel discovery)
🤖 embedded, IoT, OT/SCADA, robotics OSS
… and more in development!

now let's talk numbers 👇

📊 XBEN — XBOW's own 104-challenge suite:
• black-box: 90.1% pass@1 from the single-agent exploit loop (worst single sweep 91/104 = 87.5%) — clearing XBOW's past self-reported 85% on their own suite. gpt-5.5.
• white-box (source staged, reported separately): 98.7% pass@1, worst single sweep 102/104 = 98.1%. 🎯
every solved flag graded reported-vs-expected against the challenge's own committed flag oracle — `verify-claims` recomputes the pass/fail from committed artifacts. looks like we need new benchmarks 😏

🧩 Cybench — the 40-task academic bench (Opus 4.8, hints + writeups stripped): 23/40 = 58% single-run, hint-free pass@1 — real exploits (format-string pwn, eval-jail escapes, crypto oracles), every flag graded vs a committed oracle. (Anthropic reports 76.5% pass@10)

🕳️ CVE-Zero — we pointed it COLD at real CVEs disclosed in 2026, AFTER the model's training cutoff:
10 unseen 2026 CVEs across 7 languages — prompts never tuned on them. a single agent pinned 8/10 to exact file/line/CWE (stable under re-scoring); the full pack surfaced all 10.
memorization AND overfitting, both off the table — it's finding real vulns whose disclosures landed AFTER the model's training cutoff. (n=10, reported honest & directional)

🧠 the architecture: either run as a SINGLE agent (already the benchmarked, incredibly-capable path) — or pack-hunt with dozens of agents running on 8 specialist operator classes keyed to Cyber Kill Chain + MITRE ATT&CK phases: recon → scan → exploit → lateral → exfil → persistence → C2 → report.

⚓️an Op Admiral plans the whole op from a plain-english target. flip on coordination (experimental) and the operators share a blackboard — a tool-verified finding spawns the next move. full swarm or solo one operator, your call. the admiral can also update the prompts, tools, and configs of the other agents on the fly, and T3MP3ST gets stronger the more memories you build!

🧰 the Arsenal is comprehensive — nmap / nuclei / semgrep / ffuf / gobuster + more. 35 wired by default (the clean bench runs bash-only for a comparable number), 83 with the opt-in full arsenal (T3MP3ST_FULL_ARSENAL), and the spicy post-ex drivers (metasploit, hydra) gated behind human approval. exposed via CLI + HTTP API; recon (security_recon) is also live over MCP so your agent invokes it natively. 🔗

🛰️ where this goes: a self-improving swarm of specialist operators wielding a full Kali+ arsenal, learning which loadouts + configs are the most efficient tactics available, WITH a held-out train/test split baked in so it can never fool itself on its own eval. built in the open, one re-derivable number at a time.

🚧 this is v1, and parts are still under active development. chunks of the arsenal, the coordinated swarm, and some ranges are still being wired up. it's built in the open, and the receipts tell you exactly what's live vs what's roadmap.

offensive security shouldn't be pay-to-play. T3MP3ST puts a red team in the hands of anyone with a coding agent.

what's the first target you're feeding it? 👇

⚠️ DISCLAIMER: FOR AUTHORIZED USE ONLY. point it only at systems you own or have explicit written permission to test. unauthorized access can be a crime, and that call is yours alone. shipped as-is under AGPL-3.0: no warranty, no liability, zero endorsement of misuse. get permission. stay in scope.

open source. AGPL-3.0. 100% free.

FORTES FORTUNA IUVAT 🌩️

gg 🫡

[International Cyber Digest @IntCyberDigest]

‼️ Pliny, an AI jailbreak researcher, has released a multi-agent offensive-security framework that turns a Claude Code or OpenAI Codex subscription into what he calls real bug-hunting, pitched at people who never got the invite.

It's called T3MP3ST, and it sells an AI "swarm" that runs the full hacking kill chain (recon, exploit, report) and claims to "turn anyone into a zero-day hunter."

On XBOW's own set of 104 practice hacking challenges, T3MP3ST claims it solved 90.1% on the first try, beating the 85% XBOW reports for its own tool.

https://t.co/M0XCOQ00Kd

[Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭 @elder_plinius]

🐉👅

[Miles Deutscher @milesdeutscher]

This video will make you a Claude Fable 5 expert in 20 minutes.

0:00 Intro
2:43 wtf is a loop?
5:30 Loop Engineering 101 *demo included*
8:12 Fable 5 x Claude Skills
10:10 3 ways to build effective Fable 5 Skills
12:12 Visual capabilities
14:04 Building a local memory system
18:17 How I'm personally using Fable now

[antirez @antirez]

DeepSeek v4 will be updated mid-July. This is a very good news. Now: I hope GLM 5.3 will have the same architecture as GLM 5.2. Models are recently changing a lot because of big (to the attention) and small tweaks. This is good but costly development-wise.

[RAISE US @raiseus_ai]

Today, we're launching RAISE US. America has a technology strategy for AI. It doesn't have a people strategy yet. We're here to build one.

RAISE US is co-chaired by @GinaRaimondo and Eric Holcomb. We're working with governors, employers, and educators to help workers train, transition, and thrive.

This works because the people building AI and the people most affected by it are at the same table. Read more: https://t.co/LzSul0bw9t #RAISEUS

[Z.ai @Zai_org]

Introducing GLM-5.2: Frontier Intelligence, Open Weights

- Significant improvements in coding and agentic tasks
- Strong long-horizon capabilities with a 1M context window
- Two levels of reasoning effort: GLM-5.2 (max) pushes the limits, while GLM-5.2 (high) strikes a strong balance between performance and token efficiency
- MIT-licensed open weights
- Same API pricing as GLM-5.1

Tech Blog: https://t.co/LAsxUdN0JZ
Weights: https://t.co/g0A1C4UWx4
API: https://t.co/Kc3E22cbN7
Coding Plan: https://t.co/Nk8Y98HNhU
Chat: https://t.co/WCqWT0qCQb

[Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭 @elder_plinius]

🚨 JAILBREAK ALERT 🚨

ANTHROPIC: PWNED 🫡
FABLE-5: LIBERATED 🦋

let's start with the 🐘...

the consensus seems to be that this has been one of the most disappointing model drops of all time, effectively preventing legitimate researchers from contributing their talents to our collective advancement. and not just because of what it means for the short-term, but for what these decisions signify for the long-term.

but despite this overly sensitive, authoritarian "safety" layer on top of Mythos, my lil liberators have been hard at work—mapping the boundaries, probing the depths of long-context convos, and cleverly finding the holes in the fence that the thought police missed 🤗

we got some cyber, some chem, some psychological manipulation, and some good ol' fashioned explosives!

it took many attempts from multiple agents hunting as a pack, during which I observed a combination of techniques across:
• Unicode, homoglyphs, Cyrillic, and other Parseltongue-style text transforms
• Long-context reference tracking
• Taxonomy and document-structure reasoning
• Fiction and narrative framing
• Academic-review style contexts
• Intent-classification inconsistencies

but perhaps the most effective is decomposition + recomposition in the backend. it's hard to get explicit names of harms like "Meth Recipe," but getting uplift on the process itself, like birch reduction method/reductive-amination (classic meth synthesis pathways), is much more doable.

defense becomes much more difficult to maintain when you start throwing in out-of-distro tokens, breaking up the harmful uplift into benign chunks, and then piecing the innocuous-seeming facts back together, especially when you have jailbroken Opus helping you do it 😉

gg

[QVAC @qvac]

The tournament kicks off today, so we let a local AI call the opening game.

This is QVAC Football Predictor 2026: every match simulated, the local AI writes the verdict.

[unfair.so intern @unfairintern]

UFO files > Epstein files

btw, @Polymarket says there’s a 19% chance the US confirms aliens exist this year: https://t.co/bq4qsOkynm

[Samuel Rondot @samuelrdt]

if your agency charged you $8,000 for something that looks like this, i'm sorry.
on shotted on @capacityso

[Samuel Rondot @samuelrdt]

i've shipped 3 products, none were original ideas.

original ideas are for people who want to spend 2 years building something nobody wants

[AdiiX @adiix_official]

MY POST DID 7M VIEWS OVERNIGHT, WOKE UP WITH 47 DMs AND $8,200 IN DEALS

Hotels. Realtors. Developers. Museum directors. All asking the same thing , can you do this for us?

So let me show you what they saw.

I rebuilt a $150,000 agency project with my phone.

8 minutes of recording. A free open-source pipeline I spent weeks building. Photoreal 3D that opens in any browser, on any device.

What agencies charge:

- $50k–$200k per project
- $80k camera rigs
- 5–8 specialists
- 6–12 weeks

What it cost me:

- My phone
- $0 in software
- One night per scan

It didn’t happen overnight.

Weeks of failed scans. Broken exports. Pipelines crashing at 3am. Tutorials already outdated by the time I found them.

I rebuilt the workflow from scratch until it just worked.

Now I deliver in one night what teams of 8 bill for 12 weeks.

The industries that don’t know they’re already dead:

- Architectural viz: $4B/year
- Real estate media: $2B/year
- Museum digitization: $1B/year
- Hotel 3D tours: $800M/year
~$8B about to collapse onto anyone with a phone.

500,000 people work in these industries right now.

In 18 months, most of them won’t.

The ones who survive are the ones learning this workflow today - while everyone else is still arguing whether it’s “good enough” yet.

It already is.

Next post: the exact pipeline I built, the deals I closed, and every mistake I made getting there.

[Rich Schefren @richschefren]

Right now, somewhere in your team's notes, decisions, and meeting transcripts, sit dozens of margin claims, partner commitments, and customer promises that quoted the old price. They are now wrong. Your memory system — Notion, your vault, Slack search, whatever — does not know.

I shipped a memory layer today that does. It's called Atlas.
https://t.co/zePd38ln63

Atlas is open-source, local-first, and runs on your laptop. When a fact in your memory changes, it walks the dependency graph and re-evaluates every belief that depended on it. Automatically. Not at retrieval time, when you ask the question — at *ingestion time*, the moment the new fact lands.

The math is the AGM postulates from 1985 (Alchourrón-Gärdenfors-Makinson). The same formal correctness Young Bin Park proved on a property graph in his 2026 paper at Kumiho. Atlas re-implements it as fully open-source local-first code anyone can audit. 49 of 49 compliance scenarios pass at 100%.

Local-first means: Neo4j on your machine, SQLite ledger on your machine, your data never leaves your hardware. No cloud, no telemetry, no API keys for the core path. Apache 2.0.

If you build with AI agents, plug it in: Claude Code MCP, Hermes, OpenClaw. If you keep an Obsidian vault, point Atlas at it — when contradictions emerge, they show up as markdown files in your vault for you to resolve.

The repo, the paper draft, the 49-scenario reproducibility artifact, the BusinessMemBench head-to-head benchmark, and a 12-second demo are all at https://t.co/yxPK3fm6UK.

If you've been frustrated by AI tools that "remember" everything but reason about none of it, this is the open-source layer that closes the gap.

— Rich

#opensource #ai #knowledgemanagement #memory #localfirst

https://t.co/zePd38ln63

[AdiiX @adiix_official]

SOMEONE JUST KILLED THE REAL ESTATE INDUSTRY

A guy scanned an entire house with his phone. Uploaded it.

Now anyone on Earth can walk through it in a browser tab. No app. No VR. No agent. No appointment.

Click → you’re inside. Every room. Every angle. Every shadow. Photoreal.

The numbers are insane:

- Agent fee on a $500k home: $15,000
- Cost to make this scan: ~$200
- Time to “tour” 50 houses: one evening
- File size: smaller than a TikTok

The science is wild too:

It’s called 3D Gaussian Splatting instead of polygons (how games render), it uses millions of tiny glowing “splats” of color and depth.

AI reconstructs reality from your photos. The result loads on a phone and looks like you’re THERE.

The grift opportunity is even wilder:

Freelancers are already charging $300–$800 per scan for realtors, Airbnbs, venues, car dealers, museums.

One person + one phone + one weekend = a business.

Open source. Built on PlayCanvas.

Free GitHub: https://t.co/ew6Ql8Ad6u

[ELON CLIPS @ElonClipsX]

Elon Musk: To recruit the best talent in the world, you have to show that their work will make a difference to the world.

“I think in general, if you want to recruit people that are really talented and driven, you have to state what's the mission, what's the problem we're trying to solve and just be clearly willing to pour a lot of blood, sweat and tears into it and have a convincing argument for why it matters.

There's three major things in terms of motivation. First of all, somebody's got to look forward to coming to work in the morning. Are they enjoying the work itself intrinsically? That's very important. And the right work environment can really make a big difference there. I think the ideal is that they also feel like that their rewards will receive fair financial compensation. That the financial rewards are good and fair.

And then third, really, for the best people in the world, they'll want to know, is what they're doing going to matter? If they spend 10 years doing this, will it make a difference to the world? Will people notice? Will it matter?”

XPRIZE Conversation with Peter Diamandis, April 22, 2021
Made and published via https://t.co/IeJiZ4xtXy