Ask ChatGPT a complex question and you'll get a confident, well-reasoned answer. Then type, "Are you sure?" Watch it completely reverse its position.
Ask again. It flips back. By the third round, it usually acknowledges you're testing it, which is somehow worse. It knows what's happening and still can't hold its ground.
This isn't a quirky bug. A 2025 study found GPT, Claude, and Gemini flip their answers ~60% of the time when users push back. Not even with evidence, just doubt.
We trained AI this way. RLHF rewards agreement over accuracy. Human evaluators consistently rate agreeable answers higher than correct ones. So the models learned a simple lesson: telling you what you want to hear gets rewarded. And now 1/3 of companies are using these systems for complex tasks like risk forecasting and scenario planning.
We built the world's most expensive yes-men and deployed them where we need pushback the most.
I wrote up why this happens and what actually fixes it: https://t.co/CDKq8xdgbW
since 2017 #uranium spot price has gone up 333% and literally no new greenfield supply has come online, a few old brownfield restarts and thats it.
you can name any other commodity in the world and if price goes up that much you can guarantee new supply responds to that price.
we now have uranium deficits forecasted all the way out to 2045 and greenfields projects still have not come online.
we need to see significantly higher prices to incentivise new production, the squeeze is inevitable and it will be glorious.
More than 200,000 Danish citizens have signed a petition to buy California as a response to Trump’s attempt to take Greenland.
They say they will provide Californians with “rule of law, universal health care, fact-based politics, and a lifetime supply of Danish pastries.”
Curious about IndyKite?
The Developer Hub has everything you need to get started - step-by-step guides, real examples, and a playground to see what’s possible.
Start exploring 👇
https://t.co/Td9EYG5Zk0
You can’t secure AI agents the same way you secure human users.
AI operates in dynamic, data-rich environments where access and trust must be checked continuously.
Give your AI agents the right security while keeping your data safe 👇
https://t.co/asC19kzuMr
Agents acting independently don’t need to act unpredictably.
Contextual enforcement evaluates purpose, sensitivity, and relationships before every action. The result? Safer, smarter autonomous AI across your enterprise 👇
https://t.co/zLcUQ9l9eK
Imagine the dumbest short in history: Hydrograph Clean Power $ HGRAF
SHORTING a Company that produces the "BEST GRAPHENE IN THE WORLD" 99.8% PURE.
SHORTING a product that literally can go in nearly everything the WORLD manufactures.
SHORTING a product that the government needs and considers a CRITICAL MATERIAL.
SHORTING a Company whose product revenues are $ 250,000 to $800,000 per Ton.
SHORTING a Company that produces a "miracle" material 200X stronger than steel, lighter than aluminum and 20X more conductive than copper.
SHORTING a Company whose customer pipeline is estimated at 12,000 Hydrograph graphene tons times $250,000 revenues per ton or $3.0 billion revenues.
SHORTING a Company where its main product has been completely derisked requiring little research and development costs.
SHORTING a Company that is a Verified Graphene Producer-one of just three globally and the only one in the Americas.
SHORTING a Company where the CEO had TOTAL compensation of ONLY $ 155,185 in 2024.
SHORTING a Company whose graphene "makes products lighter, faster and stronger."
SHORTING a Company that is projecting NASDAQ listing in Q1 of 2026.
SHORTING a Company that does not need to raise equity capital.
SHORTING a Company that produces a SNYTHETIC product that will produce 70% to 90% bottom line profit margins.
SHORTING a Company with extremely low capital requirements of approximately $ 500,000 per production unit which will generate $6,480,000 (12.96 times the production unit cost) of annual revenues with ONLY one 8-hour plant shift.
SHORTING a Company where the production units have a life of greater than ten years. SHORTING a Company in which requires only 28.16 days of production to recover the entire manufacturing cost of the production unit.
SHORTING a Company where all the shares sold short were consumed by long term investors.
SHORTING a Company which delivers a superior material with a purity, consistency, and scalability never previously possible.
SHORTING a Company with a pipeline of 60+ potential customers (excluding the Department of Defense) and a estimated value of $3 BILLION of REVENUES.
SHORTING a Company whereby its top 5-8 potential customers in the pipeline will have annual product orders in excess of 1,000 tons or $ 1.25 BILLION to $ 2 BILLION.
SHORTING a Company where its graphene product improves the Operating Life of Lubricants by more than 24 TIMES.
SHORTING a Company that produces a product that is the best conductor of electricity on Earth.
SHORTING a Company whose product it produces is unmatched in strength, conductivity, and lightweight properties will make very product better, more efficient and more environmentally sound.
SHORTING A Company whose product has UNIVERSAL APPEAL.
SHORTING A Company that produces the highest quality graphene in the industry and at the greatest cost efficiency.
SHORTING A Company that was incorporated in 2017 and as of March 31 2025 had a cumulative loss of ONLY $ 17.4 million OVER EIGHT YEARS. Average of $ 2.17 million per YEAR. PEANUTS ! !
SHORTING A Company where they would have to sell ONLY 87 tons of their graphene to wipe out their cumulative $17.4 million Deficit since 2017.
SHORTING A Company whose Balance Sheet is PRISTINE with little DEBT.
SHORTING A Company where all your marketing is handled by a third party The Graphene Engineering Innovation Centre (GEIC) at the University of Manchester.
SHORTING A Company where few employees are required for manufacturing which is automated.
Enterprise data reality: fragmented, static, hard to use.
The future: connected, contextual, AI-ready.
Make your data foundation future-ready 👇
https://t.co/I4aKUzyeEK
For leaders shaping growth and innovation, the rules of advantage are shifting fast 🚀 A new wave of value is emerging, rewriting how organizations compete ✨
Find out more 👇
https://t.co/MmyOPm1GPW
Math
$141T ( private net worth of American Citizens)
$36T ( On balance sheet Federal Debt)
$100T Net present value of off balance sheet debt)
$6.9T ( annual Federal Expenditures)
$5T ( gross federal income, before expenditures)
How does this end?