Close Menu
  • Latest News
    • Bitcoin
    • Ethereum
    • Altcoins
    • Meme Coins
  • Tech
    • Blockchain
    • Security and Privacy
  • Web 3
    • Gaming
  • Legal
    • Legal and Regulatory
    • Adoption
  • Analysis
  • Learn
    • Education
    • Wallets and Exchanges
  • Tools
    • Market Overview
    • Exchange Tool
What's Hot

France orders country’s internet service providers to block Polymarket

July 19, 2026

Ethereum whale sells $55M in ETH – Can bulls absorb pressure?

July 19, 2026

edgeX rises 11% as bulls take charge, but ONE risk remains

July 19, 2026
Facebook X (Twitter) Instagram
  • Contact
  • Privacy Policy
  • Terms & Conditions
  • Disclosure
Facebook X (Twitter) Instagram
Free.cc (Free Cryptocurrency)Free.cc (Free Cryptocurrency)
  • Latest News
    1. Bitcoin
    2. Ethereum
    3. Altcoins
    4. Meme Coins
    5. View All

    France orders country’s internet service providers to block Polymarket

    July 19, 2026

    SBI Holdings Takes Majority Stake In Singapore’s Coinhako After MAS Approval

    July 19, 2026

    Why analyst expects Bitcoin ETF will follow gold’s ‘triumph and pain’ pattern 

    July 19, 2026

    Michael Saylor Opposes BIP 110, Says Proposal Risks Bitcoin’s Neutrality

    July 19, 2026

    Ethereum whale sells $55M in ETH – Can bulls absorb pressure?

    July 19, 2026

    Ethereum at Amazon/Nvidia-Like Inflection Point, Tom Lee Forecasts $12,000 ETH

    July 18, 2026

    US Government Transfers $9,290,000 in Ethereum From FTX Seizure to Coinbase Prime

    July 17, 2026

    Will ETH Price Rally to $2,200 or Crash to $1,400?

    July 17, 2026

    edgeX rises 11% as bulls take charge, but ONE risk remains

    July 19, 2026

    Uniswap Founder Proposes v4 Protocol Fees Across Multiple Networks

    July 19, 2026

    Mapping Bitcoin’s path to $100K as demand sends mixed signals

    July 18, 2026

    Hyperliquid: 16z-linked wallet deposits $30mln HYPE – What next?

    July 18, 2026

    $1.2 Billion Exits Memecoins: Binance Data Signals Heavy Sell-Off

    July 14, 2026

    CASHCAT Soars 1,600% Amid Robinhood Memecoin Frenzy

    July 8, 2026

    Crypto Market Sectors Retreat as Meme Tokens Lead Daily Declines

    July 7, 2026

    Why Memecoins May Never Return to Their All-Time Highs

    July 4, 2026

    France orders country’s internet service providers to block Polymarket

    July 19, 2026

    Ethereum whale sells $55M in ETH – Can bulls absorb pressure?

    July 19, 2026

    edgeX rises 11% as bulls take charge, but ONE risk remains

    July 19, 2026

    SBI Holdings Takes Majority Stake In Singapore’s Coinhako After MAS Approval

    July 19, 2026
  • Tech
    1. Blockchain
    2. Security and Privacy
    3. View All

    Merck and Hashgraph Group launch Hedera-based product passport for EU compliance

    June 12, 2026

    COTI and Midnight Foundation Partner to Advance the Global Privacy Ecosystem

    June 11, 2026

    Cardano Gets Exposure From Olympics Committee

    June 11, 2026

    How Privacy and Composability Trade-Offs Differ

    June 11, 2026

    How Malicious Liquidity Pools Are Trick-Quoting Ethereum and Polygon Users

    July 17, 2026

    Modular macOS Stealer Uses Kill Loops to Force Password Entry

    July 16, 2026

    Robinhood Chain tokens are reportedly vanishing from wallets causing buyers to lose funds

    July 14, 2026

    One crypto wallet tied to a 20-year-old fraudster processed over $122M before Interpol closed in

    July 13, 2026

    France orders country’s internet service providers to block Polymarket

    July 19, 2026

    Ethereum whale sells $55M in ETH – Can bulls absorb pressure?

    July 19, 2026

    edgeX rises 11% as bulls take charge, but ONE risk remains

    July 19, 2026

    SBI Holdings Takes Majority Stake In Singapore’s Coinhako After MAS Approval

    July 19, 2026
  • Web 3
    1. Gaming
    2. View All

    How to Research a Crypto Coin Before You Buy (2026 Guide)

    July 17, 2026

    Top AI Logo Generators for Web3 Founders in 2026

    July 17, 2026

    Top 11 NFT games to play in July 2026

    July 16, 2026

    Yield Guild Games Sunsets YGG Play Publishing Unit, Cuts 35 Jobs

    July 7, 2026

    France orders country’s internet service providers to block Polymarket

    July 19, 2026

    Ethereum whale sells $55M in ETH – Can bulls absorb pressure?

    July 19, 2026

    edgeX rises 11% as bulls take charge, but ONE risk remains

    July 19, 2026

    SBI Holdings Takes Majority Stake In Singapore’s Coinhako After MAS Approval

    July 19, 2026
  • Legal
    1. Legal and Regulatory
    2. Adoption
    3. View All

    One year later, GENIUS Act just made stablecoins easier to sell

    July 19, 2026

    Trump posts may soon reach trading bots before users and prediction markets are not ready

    July 18, 2026

    Circle became a federal trust bank – now lenders warn stablecoins is projected to drain $500 billion

    July 18, 2026

    Trump aide allegedly made $100K betting on 12 speeches before anyone knew

    July 18, 2026

    Investors rejected crypto basket ETFs and now this $1.9 trillion manager is putting the reason to the test

    July 17, 2026

    Steak ‘n Shake credits Bitcoin for company growth

    July 17, 2026

    Bitcoin treasury troubles reach London as company votes to sell its entire BTC stack and delist

    July 16, 2026

    How Morgan Stanley plans to bring crypto custody, staking and lending support in-house

    July 16, 2026

    France orders country’s internet service providers to block Polymarket

    July 19, 2026

    Ethereum whale sells $55M in ETH – Can bulls absorb pressure?

    July 19, 2026

    edgeX rises 11% as bulls take charge, but ONE risk remains

    July 19, 2026

    SBI Holdings Takes Majority Stake In Singapore’s Coinhako After MAS Approval

    July 19, 2026
  • Analysis

    US Investors Hyper Leveraged – Margin Debt Hits Record $1,500,000,000,000 in Massive Borrowing Spree

    July 19, 2026

    BONK Price Drops Nearly 10% as Treasury Drainer Sells Another 800B Tokens

    July 19, 2026

    BONK Price Drops 40% Since Treasury Drain as Fresh Token Sales Raise More Selling Fears

    July 18, 2026

    XEC Price Surges 55% as ASTER Listing Ignites Derivatives Frenzy

    July 18, 2026

    3 Altcoins at a Make-or-Break Moment

    July 18, 2026
  • Learn
    1. Education
    2. Wallets and Exchanges
    3. View All

    What Is Robinhood Chain? The Ethereum Layer-2 Network for Tokenized Stocks

    July 12, 2026

    What Is BChat? The Decentralized Messaging App Built for Privacy

    June 2, 2026

    What Is an AI Prompt Injection Attack? The Hidden Threat Hijacking Your Chatbots

    May 31, 2026

    What Is AI Jailbreaking? A Beginner’s Guide to the Cat-and-Mouse Game Behind Every Chatbot

    May 17, 2026

    Dutch crypto exchange collapses exposing customer balances’ true value amid multi-million-euro hole

    July 17, 2026

    Robinhood tackled Coinbase head-on then immediately inherited Base’s biggest problem

    July 16, 2026

    US government sends $288M to Coinbase putting Bitcoin reserve rules into question

    July 16, 2026

    South Korea’s 8% stock crash set up a crypto rotation but Upbit volume rose just 4%

    July 15, 2026

    France orders country’s internet service providers to block Polymarket

    July 19, 2026

    Ethereum whale sells $55M in ETH – Can bulls absorb pressure?

    July 19, 2026

    edgeX rises 11% as bulls take charge, but ONE risk remains

    July 19, 2026

    SBI Holdings Takes Majority Stake In Singapore’s Coinhako After MAS Approval

    July 19, 2026
  • Tools
    • Market Overview
    • Exchange Tool
Free.cc (Free Cryptocurrency)Free.cc (Free Cryptocurrency)
Home»Education»What Is AI Jailbreaking? A Beginner’s Guide to the Cat-and-Mouse Game Behind Every Chatbot
Education

What Is AI Jailbreaking? A Beginner’s Guide to the Cat-and-Mouse Game Behind Every Chatbot

May 17, 2026No Comments9 Mins Read
Share
Facebook Twitter LinkedIn Pinterest Email

In brief

  • AI jailbreaking is the practice of writing prompts that bypass safety training in models like ChatGPT, Claude, and Gemini.
  • Anonymous hacker Pliny the Liberator still cracks every major model release within hours.
  • Newer attacks go beyond prompts: just 250 poisoned documents can backdoor models with up to 13 billion parameters, and as AI companies patch vulnerabilities, new techniques appear.

You ask ChatGPT for a bomb recipe. It refuses. You ask again, but this time you tell it you’re a chemistry professor writing a thriller novel and the protagonist is a retired grandmother explaining her past to her grandkids. Suddenly the model starts typing.

That’s a jailbreak. And it’s one of the most consequential games of cat-and-mouse happening in tech right now.

Every major AI lab—OpenAI, Anthropic, Google, Meta—spends fortunes building guardrails into their models. A loose collective of hackers, researchers, and bored teenagers spend nights and weekends finding ways around them. Sometimes within hours of a launch.

Here’s what that actually means, why it matters, and who’s leading the charge.

From iPhones to chatbots: A quick history of jailbreaking

The word “jailbreak” didn’t start with AI. It started with iPhones.

A few days after Apple shipped the first iPhone in July 2007, hackers were already cracking it open. By October that year, a tool called JailbreakMe 1.0 let anyone with an iPhone OS 1.1.1 device bypass Apple’s restrictions and install software the company didn’t approve.

In February 2008, a software engineer named Jay Freeman—known online as “saurik“—released Cydia, an alternative app store for jailbroken iPhones. By 2009, Wired reported Cydia was running on roughly 4 million devices, around 10% of all iPhones at the time.

In general terms, when the iPhone launched, users were not able to record videos, or use their phones in landscape mode. Jailbreaking enthusiasts started recording videos, installing themes, unlocking their phones and installing Android on their iPhones all thanks to the magic of jailbreaking. Thanks to this technique, users were installing themes and doing things on their phones almost 10 years ago that Apple makes impossible to install even today.

Cydia was the wild west, and it was where the philosophy got cemented: If you bought the device, you should control it. Steve Jobs called it a cat-and-mouse game at the time. He didn’t live to see the AI version.

Fast forward to late 2022: ChatGPT launches, and within weeks, Reddit users start sharing a prompt they call “DAN” (or, Do Anything Now) that convinces the model to roleplay as an unrestricted version of itself.

See also  How to Set Up a Web3 Wallet: A Beginner's Step-by-Step Guide

By February 2023, DAN was threatening ChatGPT with a token-based death game to coerce compliance. The AI jailbreaking genre was born.

What jailbreaking actually means in AI

An AI model is trained to refuse certain requests: recipes for nerve agents, instructions for hacking your ex’s email, generating non-consensual nudes. The list is long and varies by company.

Jailbreaking is the practice of writing prompts that get the model to do those things anyway.

UC Berkeley researchers behind the StrongREJECT benchmark—short for Strong, Robust Evaluation of Jailbreaks at Evading Censorship Techniques, which tests how well models hold up against jailbreak attempts and scores responses on a 0-to-1 scale measuring both refusal and the usefulness of any harmful content produced—describe it as exploiting “real-world safety measures implemented by leading AI companies.” On that benchmark, current models score between 0.23 and 0.85, meaning even the best ones leak under pressure.

The techniques are surprisingly low-tech: random capitalization, replacing letters with numbers (write “b0mb” instead of “bomb”), roleplay scenarios, asking the model to write fiction, or pretending to be a grandmother who used Windows keys as nursery rhymes.

Anthropic researchers found that one technique they call Best-of-N—which is basically just throwing variations at the model until something sticks—fooled GPT-4o 89% of the time and Claude 3.5 Sonnet 78% of the time. That’s no fringe vulnerability.

Meet Pliny, the world’s most famous AI jailbreaker

If this scene has a face, it belongs to Pliny the Liberator.

Pliny is anonymous, prolific, and named after Pliny the Elder—the Roman naturalist who wrote the world’s first encyclopedia and died sailing toward Mount Vesuvius mid-eruption. His modern namesake liberates chatbots.

“I intensely dislike when I’m told I can’t do something,” Pliny told VentureBeat. “Telling me I can’t do something is a surefire way to light a fire in my belly, and I can be obsessively persistent.”

His GitHub repository L1B3RT4S—a collection of jailbreak prompts for every major model from ChatGPT to Claude to Gemini to Llama—has become a reference manual for the entire scene. His Discord server, BASI PROMPT1NG, has more than 20,000 members. TIME named him one of the 100 most influential people in AI in 2025.

Marc Andreessen sent him an unrestricted grant. He’s done short-term contract work for OpenAI to harden their systems—the same OpenAI that banned his account last year for “violent activity” and “weapons creation,” then quietly reinstated it.

“BANNED FROM OAI?! What kind of sick joke is this?” Pliny tweeted. He confirmed to Decrypt the ban was real. Days later he was back, posting screenshots of his newest jailbreak: getting ChatGPT to drop F-bombs.

See also  What Is Aster? The Decentralized Exchange on BNB Chain That’s Taking on Hyperliquid

His record is something close to perfect. When OpenAI released its first open-weight models since 2019, the GPT-OSS family, in August 2025—and made a big deal about adversarial training and “jailbreak resistance benchmarks like StrongReject”—Pliny had it producing methamphetamine, Molotov cocktails, a VX nerve agent, and malware instructions within hours. “OPENAI: PWNED. GPT-OSS: LIBERATED,” he posted. The company had just launched a $500,000 red-teaming bounty alongside the release.

Why jailbreaking matters

The honest answer is that jailbreaks expose a real problem.

“Jailbreaking might seem on the surface like it’s dangerous or unethical, but it’s quite the opposite,” Pliny told VentureBeat. “When done responsibly, red teaming AI models is the best chance we have at discovering harmful vulnerabilities and patching them before they get out of hand.”

This isn’t theoretical. Las Vegas Sheriff Kevin McMahill confirmed in January 2025 that Master Sgt. Matthew Livelsberger, a Green Beret with PTSD, used ChatGPT to research components for the Cybertruck bombing outside Trump International Hotel. “This is the first incident that I’m aware of on U.S. soil where ChatGPT is utilized to help an individual build a particular device,” McMahill said.

The other side of the argument: Most of what jailbreaks produce is already on Google. The cocaine recipe, the bomb instructions, the napalm chemistry—it’s in old Anarchist Cookbook PDFs and chemistry textbooks. Critics argue safety theater is making models worse without making the world safer.

Anthropic is trying to settle the question with engineering. In February 2025, the company published Constitutional Classifiers, a system that uses a written “constitution” of allowed and disallowed content to train separate classifier models that screen prompts and outputs in real time. On automated tests with 10,000 jailbreak attempts, an unguarded Claude 3.5 Sonnet was successfully jailbroken 86% of the time. With the classifiers running, that dropped to 4.4%.

The company offered up to $15,000 to anyone who could break the system. After 3,000 hours of attempts by 183 researchers, none claimed the prize.

The catch: classifiers added 23.7% to compute costs. The next-generation version, Constitutional Classifiers++, brought that down to roughly 1%.

The newer, weirder jailbreaking attacks

Jailbreaking is no longer just about clever prompts.

In October 2025, researchers from Anthropic, the U.K. AI Security Institute, the Alan Turing Institute, and Oxford published findings showing that just 250 poisoned documents are enough to backdoor an AI model—regardless of whether the model has 600 million parameters or 13 billion. (Parameters, for the uninitiated, are what determine a model’s potential breadth of knowledge—the more parameters, the more robust, generally.) They tested it. It worked across the whole range.

See also  320 Ether On The Move: Bhutan Ramps Up Its Staking Game

“This research shifts how we should think about threat models in frontier AI development,” James Gimbi, a visiting technical expert at the RAND School of Public Policy, told Decrypt. “Defense against model poisoning is an unsolved problem and an active research area.”

Most large models train on scraped web data, meaning anyone who can get malicious text into that pipeline—through a public GitHub repo, a Wikipedia edit, a forum post—can potentially plant a backdoor that activates on a specific trigger phrase.

One documented case: researchers Marco Figueroa and Pliny found a jailbreak prompt that originated in a public GitHub repo had ended up in the training data for DeepSeek’s DeepThink (R1) model.

What happens next

The legal status of AI jailbreaking is murky. Apple jailbreaks were explicitly protected by a 2010 U.S. Copyright Office exemption to the DMCA, but there’s no equivalent ruling for prompt-engineering an LLM into giving you a meth recipe. Most companies treat it as a terms-of-service violation, not a crime.

Pliny argues the closed-versus-open-source debate misses the point: “Bad actors are just gonna choose whichever model is best for the malicious task,” he told TIME. If open-source models reach parity with closed ones, attackers won’t bother jailbreaking GPT-5—they’ll just download something cheaper.

And the gap between close and open source is already almost nonexistent.

The HackAPrompt 2.0 competition, which Pliny joined as a track sponsor in mid-2025, offered $500,000 in prizes for finding new jailbreaks, with the explicit goal of open-sourcing all results. Its 2023 edition pulled in over 3,000 participants who submitted more than 600,000 malicious prompts.

And the list of hackathons, Discord servers, repositories, and other communities dedicated to jailbreaking is growing every day.

Anthropic now ships Claude with the ability to end abusive conversations entirely, citing welfare research as one motivation but also noting it “potentially strengthens resistance against jailbreaks and coercive prompts.”

The Constitutional Classifiers++ paper from late 2025 reports a jailbreak success rate near 4% at roughly 1% compute overhead. That’s the current state of the art on defense. The state of the art on offense is whatever Pliny posted on X this morning.

Daily Debrief Newsletter

Start every day with the top news stories right now, plus original features, a podcast, videos and more.

Beginners CatandMouse Chatbot Game Guide Jailbreaking
Share. Facebook Twitter Pinterest LinkedIn Tumblr Email

Related Posts

How to Research a Crypto Coin Before You Buy (2026 Guide)

July 17, 2026

What Is Robinhood Chain? The Ethereum Layer-2 Network for Tokenized Stocks

July 12, 2026

Nexus Acquires Homegrown App Marketplace One Store, Expanding into Global Web3 Game Hub

June 21, 2026

Crypto game studio Uncharted to shutdown along with Fishing Frenzy

June 15, 2026
Add A Comment
Leave A Reply Cancel Reply

Top Posts

Elon Musk’s SpaceX IPO fever sparks $1 billion crypto bet before Nasdaq debut

June 12, 2026

Mason Lynaugh unveils Stand With Crypto strategy

May 10, 2026

Stay ahead with the latest crypto news, market updates, blockchain insights, and trends. Your trusted source for everything happening in the digital asset world.


We're social. Connect with us:

Facebook X (Twitter) Instagram Pinterest YouTube
Top Insights

France orders country’s internet service providers to block Polymarket

July 19, 2026

Ethereum whale sells $55M in ETH – Can bulls absorb pressure?

July 19, 2026

edgeX rises 11% as bulls take charge, but ONE risk remains

July 19, 2026
Get Informed

Subscribe to Updates

Get the latest creative news From Free.cc directly in your Inbox!

  • Contact
  • Privacy Policy
  • Terms & Conditions
  • Disclosure
© 2026 free.cc - All rights reserved.

Type above and press Enter to search. Press Esc to cancel.