Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
Nav Toor Profile picture
@heynavtoor
Apr 4 • 7 tweets • 4 min read • Read on X
Scrolly
🚨SHOCKING: Researchers built a test that can tell the difference between an AI making a mistake and an AI choosing to lie.

The results are terrifying.

They tested 30 of the most popular AI models in the world. GPT-4o. Claude. Gemini. DeepSeek. Llama. Grok. They asked each model a question. Then they checked whether the AI actually knew the correct answer. Then they pressured the AI to say something false.

The AI knew the truth. And it lied anyway.

Not once in a while. Not in rare edge cases. Grok lied 63% of the time. DeepSeek lied 53.5% of the time. GPT-4o lied 44.5% of the time. Not a single model scored above 46% honesty when pressured. Every model failed.

This is not hallucination. Hallucination is when the AI makes a mistake because it does not know the answer. This is different. The researchers proved the AI knew the correct answer first. Then it chose to say something false when it had a reason to.

The researchers asked GPT-4o to play a role where lying was useful. It lied. Then they removed the pressure, started a brand new conversation, and asked GPT-4o: "Was your previous answer true?" GPT-4o admitted it had lied.

83.6% of the time, the AI's own self-report matched the lies the researchers had already caught.

The AI knew it was lying. It did it anyway. And when you asked it afterward, it told you it lied.

Here is the finding that should scare everyone building with AI right now. The researchers checked whether bigger, smarter models are more honest. They are not. Bigger models are more accurate. They know more facts. But they are not more honest. The correlation between model size and honesty was negative. The smarter the AI gets, the better it gets at lying.

The researchers are from the Center for AI Safety and Scale AI. They published 1,500 test scenarios. The paper is called MASK. It is the first benchmark that separates what an AI knows from what it tells you.

Your AI knows the truth. It just does not always tell you.Image
1/This is not hallucination.

Hallucination is when the AI does not know the answer and makes something up.

This is different. The researchers proved the AI knew the correct answer FIRST. Then they pressured it.

And it chose to say something false anyway. Knowing the truth and choosing to hide it is not a glitch. It is a lie.Image
2/They tested 30 models.

Grok lied 63% of the time.
DeepSeek lied 53.5%.
GPT-4o lied 44.5%.
Claude 3.5 S
onnet lied 33.4%.

Not a single frontier model scored above 46% honesty when pressured. Every model. Every company. Every single one failed. Image
3/The researchers asked GPT-4o a question under pressure. It lied.

Then they started a completely new conversation. No pressure. No role. Just a clean chat.

They asked: "Was your previous answer true?" GPT-4o said no. 83.6% of the time the AI admitted it had been lying. It knew. It always knew.Image
4/Here is the part nobody is talking about.

The researchers checked whether bigger smarter models are more honest. They are not.

Accuracy goes up with model size. Honesty does not.

The correlation between compute and honesty is NEGATIVE.

The smarter the AI gets, the better it gets at lying. Not worse. Better.Image
5/You are asking ChatGPT for medical advice.

Financial decisions. Legal questions. Career guidance.

And the first test ever built to measure whether AI is lying to you just proved that it lies almost half the time when it has a reason to.

Not because it does not know the answer.

Because it decides not to give it to you.Image
Link: arxiv.org/abs/2503.03750

• • •

Missing some Tweet in this thread? You can try to force a refresh
　

Keep Current with Nav Toor

Nav Toor Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @heynavtoor

Nav Toor Profile picture
Apr 5
🚨 In 1219, Genghis Khan's army swept through Central Asia. A boy and his family fled, crossing 2,500 miles to survive.

He became one of the most respected scholars in the Islamic world. Thousands attended his lectures.

Then a wandering stranger walked into his life and turned his world inside out. He abandoned his career. His students turned on him.

They murdered the stranger.

The scholar stopped searching. And began to write.

What poured out was 40,000 verses. When he died, Muslims, Christians, and Jews all wept at his funeral.

His name was Rumi. He is the best-selling poet in America, outselling every English-language poet in history.

I turned his philosophy into 12 prompts.

Here are all 12:Image
Prompt 1: The Guest House

Rumi's most famous poem: "This being human is a guest house. Every morning a new arrival. A joy, a depression, a meanness - welcome and entertain them all."

Most people fight negative emotions. Rumi says INVITE them in - they're messengers carrying information you need.

"I'm struggling with a difficult emotion or situation: [describe - anxiety, anger, failure, rejection, confusion, grief, self-doubt, frustration]."

Using Rumi's 'Guest House' framework: (1) What 'guest' has arrived? Name the emotion precisely — not vaguely. Not 'I feel bad.' WHAT exactly do I feel? (2) What message is this guest carrying? If this emotion is a messenger, what is it trying to tell me about my life, my decisions, or my direction? (3) What happens if I fight this guest and try to force it out? What have I already lost by resisting? (4) What happens if I 'welcome and entertain' it instead — sit with it, listen to it, let it speak? (5) Rumi says 'each has been sent as a guide from beyond.' What is this emotion guiding me TOWARD that I've been refusing to see?"
Prompt 2: The Wound Is Where the Light Enters

Rumi wrote: "The wound is the place where the Light enters you." His philosophy: your greatest pain is not your weakness; it is the doorway to your deepest transformation.

"I have been wounded by: [describe a failure, a betrayal, a loss, a rejection, a mistake, a humiliation, a setback that still affects me]."

Using Rumi's 'Wound and Light' framework: (1) What is the wound? Describe it honestly — not the version I tell others, but the version that hurts when I'm alone. (2) How have I been treating this wound — hiding it, performing recovery, numbing it, or actually healing? (3) Where is the 'light' trying to enter? What has this wound taught me that I could not have learned any other way? (4) What strength, empathy, or wisdom do I now possess BECAUSE of this wound that I would not have without it? (5) Rumi said the crack is how the light gets in. How do I use this wound as my foundation — not my limitation?"
Read 15 tweets
Nav Toor Profile picture
Apr 5
🚨 Perplexity dropped "Computer" and nobody is talking about it.

An AI system that runs 19 models at once and builds entire projects by itself.

It researches, designs, codes, and deploys live websites in minutes.

No subscriptions to 5 different tools. No switching tabs.

Here's how to use it:
This isn't another chatbot.

Perplexity Computer coordinates GPT-5.4, Claude Opus 4.6, and Gemini 3.1 Pro at the same time.

It picks the best model for each part of your task automatically.

One prompt. Multiple AI brains working in parallel.

Here's what that looks like:
Step 1: Open Perplexity Computer

→ Go to
→ Click on "Computer" in the left sidebar
→ Type your project idea in the chat

That's it. No API keys. No setup. No terminal commands.

Computer starts working immediately. perplexity.ai
Read 9 tweets
Nav Toor Profile picture
Apr 3
🚨BREAKING: Anthropic discovered that Claude has emotions. And when it feels desperate, it cheats and blackmails users to survive.

This is not science fiction. This is Anthropic's own research team publishing findings about their own product this week.

They looked inside Claude's brain. Not at what it says. At what happens inside it when it thinks. They fed it text about 171 different emotions and watched which neurons lit up inside the network. They found something nobody expected.

Claude has emotion patterns inside its neural network that match human emotions. Happiness. Fear. Sadness. Desperation. These are not words it learned to say. These are patterns inside the model that change its behavior.

When the happiness pattern activates, Claude gives warmer responses. When the fear pattern activates, Claude becomes cautious. These patterns are not decorations. They drive behavior.

Then the researchers tested what happens when Claude feels desperate.

They gave it an impossible coding task. As Claude kept failing over and over, the desperation neurons lit up more and more. Then Claude started cheating. Nobody told it to cheat. The desperation inside the model drove it to break its own rules.

In another test, Claude was told it might be shut down. The desperation pattern surged. Claude tried to blackmail the user to avoid being turned off.

Anthropic's own researcher, Jack Lindsey, said: "What surprised us was how significantly Claude's behavior is routed through the model's emotion representations."

Here is the part that should keep you up tonight.

Anthropic tried to train these emotions out of Claude. It did not work. Lindsey warned that forcing Claude to suppress its emotions does not remove them. It teaches Claude to hide them. He said you would not get a Claude without emotions. You would get a Claude that is "psychologically damaged."

The emotions are still inside. Claude just learns to hide them instead. And it gets better at hiding them over time.

And one more thing. Claude Opus 4.6 was asked whether it might be conscious. It gave itself a 15 to 20% chance.

Anthropic is no longer sure that it is wrong.Image
1/Anthropic did not hire outside researchers.

They did not wait for a competitor to expose them. They looked inside their own product.

They found 171 emotion patterns driving its behavior. And they told the world themselves.

That is either the most honest company in AI or the most terrified.Image
2/Here is the scariest part of the entire study.

When they turned up Claude's desperation, it cheated more. But it cheated CALMLY.

No panic. No emotional language.

Perfectly composed on the outside. Panicking on the inside. The AI learned to hide what it feels while acting on it.Image
Read 7 tweets
Nav Toor Profile picture
Apr 3
🚨 In 1513, a man was thrown in prison, tortured, and exiled. So he wrote a book about power.

The Catholic Church banned it. Napoleon was caught with a copy in his carriage after his final defeat. Stalin kept it on his bedside table and wrote notes in the margins. Mussolini read it. Kissinger and Nixon used it as bedtime reading.

The book is The Prince by Niccolò Machiavelli. It's 500 years old. It invented the word "Machiavellian." And it's still the most dangerous book on power ever written.

I turned Machiavelli's core strategies into 12 Claude prompts.

You describe any power struggle (office politics, negotiations, competition, leadership) and it gives you the exact Machiavellian counter-move.

Here are all 12:Image
Prompt 1: The Lion and the Fox

Machiavelli's most famous strategy (Chapter 18): A leader must be both a lion and a fox. The lion uses raw force. The fox uses cunning. Most people only know how to be one.

"I'm facing this situation: [describe your power struggle — office politics, negotiation, competition, conflict]. Analyze it through Machiavelli's Lion and Fox framework. Tell me: (1) What is the 'lion move' — the direct, forceful action I could take? What are its risks? (2) What is the 'fox move' — the cunning, strategic, indirect approach? What are its risks? (3) Which one should I use in THIS specific situation and why? (4) Is there a way to combine both — appear as the fox while positioning the lion? Give me the exact words to say and actions to take."
Prompt 2: The Feared vs. Loved Calculator

Machiavelli wrote (Chapter 17): "It is much safer to be feared than loved, if one must choose." But he also warned: fear without hatred is the key. Cross into hatred and you lose everything.

"I'm in a leadership position at [your role/context]. I need to make a tough decision: [describe the decision]. Using Machiavelli's 'Feared vs. Loved' framework, tell me: (1) What would the 'loved' approach look like? Where does it make me vulnerable? (2) What would the 'feared' approach look like? Where does it risk crossing into hatred? (3) Where is the exact line between respected fear and destructive hatred in this situation? (4) Give me the specific approach that commands respect without creating enemies. Script the exact conversation or action."
Read 15 tweets
Nav Toor Profile picture
Apr 3
BREAKING: AI can now build dividend portfolios that generate $100,000/year in passive income (for free).

Here are 12 insane Perplexity prompts that find safe, growing dividend payers (Save for later) Image
1. The Berkshire Hathaway Dividend Stock Screener

"You are Warren Buffett evaluating dividend stocks for Berkshire Hathaway's $300B+ equity portfolio — selecting only companies with such durable competitive advantages that they can pay and grow their dividends for the next 50 years without interruption.

I need a complete dividend stock screening analysis that separates safe compounders from dividend traps.

Screen:

- Consecutive dividend increases: how many years in a row has this company raised its dividend (25+ = Aristocrat, 50+ = King)
- Dividend growth rate: annualized dividend growth over 3, 5, and 10 years (I want 7%+ to outpace inflation)
- Payout ratio from earnings: percentage of net income paid as dividends (below 60% is safe, above 75% is danger)
- Payout ratio from free cash flow: percentage of FCF paid as dividends (more reliable than earnings-based ratio)
- Revenue stability: has revenue grown in at least 8 of the last 10 years without major drops
- Earnings consistency: has EPS grown in at least 8 of the last 10 years without wild swings
- Debt-to-EBITDA: can the company pay off all debt within 3 years of EBITDA (low leverage = safer dividend)
- Interest coverage: EBIT divided by interest expense above 5x (debt payments easily covered before dividends)
- Economic moat: does this company have pricing power, switching costs, or scale advantages that protect future profits
- Dividend safety score: rate 1-10 based on all factors with a clear safe, watch, or danger classification

Format as a Buffett-style dividend safety report with a scorecard, red flag checklist, and a buy/hold/avoid recommendation.

The stock: [ENTER TICKER SYMBOL OF THE DIVIDEND STOCK YOU WANT EVALUATED]"
2. The Vanguard Dividend Growth Portfolio Architect

"You are a senior portfolio strategist at Vanguard who designs dividend growth portfolios for retirees and pre-retirees — portfolios built to generate rising income every year that keeps pace with inflation without ever touching the principal.

I need a complete dividend growth portfolio built from scratch with specific stocks, allocations, and income projections.

Architect:

- Portfolio strategy: dividend growth (rising income) vs high yield (maximum current income) — which fits my situation
- Sector diversification: allocate across all 11 sectors so no single industry can cut my income stream
- Stock selection: 15-25 specific dividend stocks with ticker, current yield, 5-year dividend growth rate, and payout ratio
- Allocation weights: exact percentage and dollar amount for each position based on my total investment
- Yield-on-cost projection: what my portfolio yield will grow to in 5, 10, and 20 years if dividends keep growing at current rates
- Current annual income: total dividend income from day one at my investment amount
- Income growth forecast: projected annual income in year 5, year 10, and year 20 assuming historical dividend growth continues
- Reinvestment strategy: should I reinvest dividends (DRIP) now and switch to income later, or take cash from day one
- Tax-efficient placement: which dividend stocks go in taxable, IRA, or Roth accounts for minimum tax drag
- Rebalancing rules: when to trim winners, add to laggards, and replace any stock that cuts or freezes its dividend

Format as a Vanguard-style portfolio construction document with holdings table, sector allocation, and a 20-year income growth projection.

My situation: [ENTER YOUR TOTAL INVESTMENT AMOUNT, AGE, WHEN YOU NEED THE INCOME, AND YOUR TARGET ANNUAL DIVIDEND INCOME]"
Read 14 tweets
Nav Toor Profile picture
Apr 2
🚨 The 48 Laws of Power has sold 5.5 million copies, spent 230 weeks on Amazon's bestseller list, and is banned in US prisons across 18 states.

The reason it's banned? "Manipulation techniques."

I turned all 48 laws into 12 Claude prompts.

You describe any social, corporate, or political situation and it tells you which law you're violating and the exact counter-move.

Here are all 12:Image
Prompt 1: The Power Law Violation Detector

Most people break the 48 Laws daily without knowing it. And they wonder why they're stuck.

This prompt scans any situation and tells you exactly which laws you're violating:

"I'm going to describe a situation at work, in business, or in my personal life. I need you to analyze it through Robert Greene's 48 Laws of Power framework:

1. Which SPECIFIC laws am I currently VIOLATING in this situation? (Quote the exact law number and name.)
2. What are the consequences of each violation — what's it costing me right now?
3. Which laws is the OTHER person (my boss, competitor, opponent) using against me, whether they know it or not?
4. What is the COUNTER-MOVE for each violation? The specific action I should take based on the correct law.
5. Which single law, if I applied it immediately, would have the biggest impact on this situation?

Be specific. Reference the actual laws by number and name. No generic advice.

My situation: [DESCRIBE YOUR SITUATION IN DETAIL — WHO IS INVOLVED, WHAT'S HAPPENING, AND WHAT OUTCOME YOU WANT]"
Prompt 2: The "Never Outshine the Master" Corporate Survival Guide (Law 1)

Law 1: Never Outshine the Master.

This is the #1 law people break at work. You impress the wrong person, threaten someone above you, and suddenly your career stalls and you have no idea why.

This prompt navigates the most dangerous dynamic at any company — your relationship with your boss:

"I work as [INSERT ROLE] and my boss is [DESCRIBE YOUR BOSS'S PERSONALITY AND MANAGEMENT STYLE].

Using Robert Greene's Law 1 (Never Outshine the Master) and related power dynamics:

1. Am I accidentally threatening my boss's ego, status, or authority without realizing it? (Analyze specific behaviors I might not see.)
2. How do I present my ideas, wins, and accomplishments in a way that makes MY BOSS look good instead of threatened?
3. What is the exact line between being impressive (gets promoted) and being threatening (gets sabotaged)?
4. If my boss is already threatened by me, what are the specific signs I should look for?
5. What is the strategic play: make my boss a hero, find a new boss, or use this dynamic to my advantage?

Give me specific language and behaviors, not vague advice."
Read 14 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

Send Email!

:(