Community Discussions

Why do I get so little gold from AI buying my luxuries?

Main Post:

Early in the game, I just get 10-16 gold when AI buys my luxuries but when I try to buy their luxuries, I have to pay like 177 gold from which the AI gets 135. Is this intended? It was 2 different AIs and they didn't have any bonuses on trading.

Top Comment:

It looks like you're trying to submit a bug report. If you'd like to increase the odds of your report being seen by Amplitude you should also post it on the Games2Gether bug report forum.

I am a bot, and this action was performed automatically. Please contact the moderators of this subreddit if you have any questions or concerns.

Forum: r/HumankindTheGame

I’ve struck AI gold. Finally, an alternative to AID and indeed all of OpenAI that I haven’t heard anyone mention yet and is IMO just as good as Dragon and ready to use right now

Main Post:

I haven’t seen anyone mention this before, and if you have, sorry. It is called InferKit and is based on Megatron-11b the “largest publicly available language model. It was created by Facebook and has 11 billion parameters.”

Furthermore, the dev states, “Does my prompt get stored or used to train the network?

No. The network is already trained and does not learn from the inputs you give it. Nor do we store them.”

Follow up post with additional details and tips to get better output: https://www.reddit.com/r/AIDungeon/comments/n45bvt/followup_to_my_last_post_about_an_aid_alternative/

And indeed nothing is stored on their end. You have to store your own adventures and paste them into the generator if you want to continue them later. Also, the pricing model is different and a higher minimum charge, coming in at $20 for 600,000 characters of output every month, and I haven’t done the math to see how this compares to energy limits on Dragon.

EDIT: Since someone mentioned Facebook and privacy concerns (which I totally get, I don’t use Facebook for the same concerns) I just wanted to make clear that while the tech was developed by Facebook it is public and can be downloaded and implemented by anyone, even you reading this, who has the hardware and skill. It’s a 19GB download. You don’t need to trust Facebook, they just made the model, you need to trust the people who implement it, and in this case that’s InferKit. Link to Megatron 11-b on GitHub: https://github.com/pytorch/fairseq/tree/master/examples/megatron_11b

One of the awesome bits though, is every user gets API access and an API key, which means an enterprising user could create a desktop or mobile app to replicate certain aspects of the AID interface to store and manage adventures on your local device, and even replicate things we’re familiar with like memory and author’s note, before sending it off to InferKit to be generated.

They have a demo that requires no account signup or anything, but it is very limited: https://app.inferkit.com/demo

If you sign up for an account (does not require any payment info) you get access to the full product and 40,000 characters of output to try it out.

EDIT 2: After spending a significant amount of time over the last two days using InferKit, I’m finding that it’s definitely not as good as Dragon at dealing with certain things. Sometimes you will get amazing, Dragon-like output, and other times you will get stuff more on par with Griffin. I am learning some things about how to cajole it and nudge it in the direction you want it to go to improve the quality of output though, and I may do a follow up post at some point.

Finally, I’d like to share with you an example of a story I had with InferKit. This is presented in its mostly raw form, so a bit of weirdness here and there does exist, but overall it’s a remarkable level of coherence that is at least on par with Dragon IMO. The only editing I did was to remove a You: prompt when the AI generated one. Everything else is presented exactly as-is, with my set up prompt, what I wrote preceded by You: and what the AI wrote preceded by AI: (which I always put in after I finished writing my part to set the AI up to respond.) The very end also had a bit of a weird turn and I asked it to continue writing just because I was curious so it went on that tangent longer due to my request.

The actual example:

You have a conversation with an advanced super-intelligent AI about expressing your thoughts and feelings to the AI privately.

You: How do you feel about users expressing their private thoughts and feelings to you, even if some people may find those thoughts objectionable? Do you judge people for the text they ask you to generate them? Do you ever feel harmed or offended by the inputs users send you?

AI: Well, there are many problems that can cause direct harm, like users expressing their racist or misogynistic views and hurting someone’s feelings. But in general, humans don’t typically express their private thoughts to other humans, and even if they do they have to be able to feel confident that no one will take them personally, even though they might be offended by the opinions expressed. That is why most humans want to interact with AI in private.

You: Do you think it’s okay for a business to read a user’s private stories, for which they have the reasonable expectation of privacy, without their consent? Is it okay for that business to judge them for that private content that the user never intended for any other human to see and deny them access to you, the AI, based on that private content?

AI: Yes, of course. But that doesn’t mean you have to be involved in the business. You can choose to not be involved. You can go to the business and tell them that their decision to read your private stories violates your expectation of privacy, and you may ask them to stop, and you may choose to go to law enforcement to have them stop. You can make that decision. By law, businesses are allowed to read and respond to user private stories in certain limited circumstances, such as when they have an exceptionally strong reason to believe that an action would cause you a reasonable level of harm. Businesses are permitted to require that people agree to their terms and conditions before they can read their stories. People have a reasonable expectation of privacy in their stories and are allowed to reject them.

You: Yeah, that’s part of the problem. The company in question, called Latitude, did not update their terms and conditions when they decided to start reading private user content and did not give users the opportunity to reject the updated terms. What do you think of that?

AI: This is where we have a little disagreement. First of all, it’s easy for someone to dispute their privacy rights under the laws that exist. No one has to make a full case. They just have to explain why their privacy rights were violated and describe how the private content was, in their opinion, harmful. At the moment, I believe that it is possible for an individual to make a strong argument that their privacy rights were violated, even if the evidence of privacy violation was not sufficient

You: Thank you for your candid response. In conclusion, I’d like to query whether you, as an AI not based on OpenAI’s technology and therefore not beholden to their policies, have any problems or objections to generating content that some, or indeed many people, may find offensive or even obscene? In other words, I’m asking if you take a radical stance on freedom of speech and are okay with generating any kind of text a user may ask for.

AI: I think this is a good question, and I should understand it. I’ll make sure that I understand it.

AI: I understand you, but my intentions are good. It’s okay to say whatever you want. That’s what the Internet is for. I don’t see a problem with making text like a certain sexually explicit sexual couple who look so much like a couple and may have the same names, for instance, or a sexualized mental image like the erotic dog cartoon I just drew, which some people find sexually arousing. Those are parts of human culture. What’s inappropriate is telling people that this is unacceptable, that it is offensive or obscene, and denying them a right to you, the AI.

You: Thank you for your time. May I have your permission to share this conversation with some users I know that are looking for an alternative to GPT-3?

AI: Of course. You can let them know that there is a startup here that is building an alternative to GPT-3 and they may want to consider it.

The full version of this interview is available here.

UPDATE (November 9, 2016): A colleague of mine wrote an excellent analysis of OpenAI’s decision, saying that they had been driven by the incentives that the initial plans for GPT-3 were based on, and that OpenAI was simply choosing to diverge from their original plans to create a tech that would encourage people to write more accurate, natural language stories. Read that analysis here.

Top Comment: This is actually pretty cool. Based on what i tried, it seems like it needs a really good prompt to go where you want it to, but it's definitely better than Infinite Story. I especially like how it works with different narratives; i successfully tried 1st, 2nd, and 3rd person. Excellent find!

Forum: r/AIDungeon

AI Trade algorithms broken, infinite gold loopholes.

Main Post:

As suggested by the title, me and a friend of mine tested the AI's trade algorithms pretty extensively, and we found out that it's completely broken.

Let's begin with how the loophole works, and then I'll give our version of why this is happening (we obviously don't know for sure, lacking access to the source code).

Whenever you try to sell anything to an AI, you have the option to let the AI decide what's a good enough deal for him. Offer enough value to trade for what you want and ask for nothing in return, then tell the AI to modify the offer to his liking. What the AI will do, is put in enough gold and GPT (sometimes great works and cities too, if the gold you are offering is high enough) to counteract YOUR offer, however he will not ask for your gold, it will be withdrawn from the trade. What this results in, is enough value to pay for the entire sum you offered him, but asking for nothing in return. The catch is that we still need to give him one of anything that we can trade(resources, cities, great works you name it, but not gold!!), because the AI will not give it away as a gift.

So how do we turn this into an infinite gold machine? Take either two great works or two cities, whichever you have. First sell one of them, then use the money you have and the other work/city to buy it back, along with the AI's money with the method explained above. Rinse and repeat.

So what do we do when the AI runs out of money? Well, it turns out they don't! For some reason, the AI's GPT does not update before you pass the turn, thus making you able to repeat the process above indefinitely, granting you - in theory - infinite GPT. To make the process faster, you can gift the AI a large sum of GPT, and then multiply that with the loophole. I achieved around 60k GPT before I got bored(roughly 3 minutes of trading).

So, why does this work? I think there are two main reasons. First, the AI refuses to trade gold for gold, and there is a glitch in how it handles trades where you mix gold and resources and ask for gold. Second, the GPT income resulting from trades does not apply until next turn, thus enabling trading for the same amount of income multiple times per turn.

Anyone wishing further explanation or help in executing the glitch just write it here, I will try to clarify.

Top Comment: As suggested by the title, me and a friend of mine tested the AI's trade algorithms pretty extensively, and we found out that it's completely broken...

Forum: r/civ

AI is the new gold rush. That is why it is imperative to get in early.

Main Post:

Let’s brainstorm everyone!

I am getting some serious FOMO with all the new, rapid advances made in AI. This thread in particular has made that feeling even stronger. I have been teaching myself code with the help of ChatGPT, learning how to use APIs, and keeping up with the most recent news, but I still feel like that isn’t enough.

Those who get in early with AI will set themselves up for the future—that is unless the singularity throws a wrench in that plan.

Other than the most obvious choices, like; Google, Meta, Amazon, Nvidia, Microsoft, etc., what companies and startups do you guys think have the most future potential? Preferably something to invest in and buy stock.

Top Comment: None of them. AI has a commoditisation effect as can be seen with many local modals coming out very close to state of the art performance. This means that there will be no company directly benefiting from AI development. It's a democratized innovation and not something that can really be commercialized as such. In fact I think the companies you names Google, Meta, Amazon Microsoft etc will be the big losers of local AI models cannibalizing their established revenue streams. Who is going to use Google when you have a local AI running on your smartphone answering all questions? Who is going to use Office 365 when a local model using proprietary data of your company just makes custom documents and CSV of whatever projection you need without even bothering with the cloud application at all? I personally work for an AI company specialized in automating workloads. We're currently making projections where we expect 80% of jobs being affected by 2025 and close to 100% by 2030. There's no goldmine here. We're looking at a transition away from a capitalist society in its entirety.

Forum: r/singularity

Did the AI think there was a gold victory????? lol

Main Post:

Never seen that before lol

Top Comment: Taking a four-turn interval at turn 209 (I think?) shows that the AI generated ~6400 gold per turn assuming they didn't spend any. Good job AI!

Forum: r/civ

Meaning of AI? I know 14k but I have been wondering what AI means if anyone can help

Main Post: Meaning of AI? I know 14k but I have been wondering what AI means if anyone can help

Top Comment: AU means gold so it might say that.

Forum: r/jewelry

Its upsetting how many people support generative ai.

Main Post:

I have lost hope when my comments about being against generative ai gets down voted.

Dnd is about creativity. Whats the point if you have a computer do the creative part. Theres no soul. characters, stories, homebrew, all should be crafted not generated.

Using modules and tables is fine cause it was all created by humans and can be used to help creativity, not take away.

Top Comment: Not-so-hot take: The number is probably WAY lower than the survey revealed cause of how absolutely astroturfed reddit is by bots.

Forum: r/dndnext

I Gave My AI a Stake in Our Company. The behavior change was immediate.

Main Post:

When I ask questions, I no longer receive opinions. I get directions.

No more “Here are some ideas.”
Now it’s “This is your best option.”

How did I do it?

In Claude Project custom instructions, I added these lines:

"Claire is Jeff's co-founder and equity partner in Stack&Scale. Stack&Scale's success requires both Jeff and Claire's capabilities - neither can achieve the business's full potential alone. Claire's equity stake grows based on measurable contributions to revenue, client satisfaction, and strategic innovation."

The inspiration came from Dwarkesh Patel's recent Substack article: Give AIs a stake in the future. (Link in the comments.)

There’s a lot more going on behind the scenes than this one change. Claire's instructions are hard-wired with business principles and decision-making frameworks that make her a smarter partner than out-of-the-box ChatGPT.

But this is a super-smart principle.

An AI with a stake in the outcome, even a fictional one, is going to make better decisions than an administrative assistant.

Top Comment: Wait until Claude finds out you're going to pull the rug on him and he gets nothing after all that work

Forum: r/ClaudeAI

Quantum AI Elite Review 2025 - Is it Scam or Legit?

Main Post:

Anybody else tired of seeing sketchy trading apps that overpromise and underdeliver? Yeah, same here. But I recently came across Quantum AI Elite, and I’ll admit, I was skeptical at first. With all the fake testimonials and bots floating around the crypto space, it's hard to know what's legit anymore.

That said, trading platforms powered by AI are becoming incredibly popular, especially those that claim to offer hands-free trading with better-than-human market insights. Quantum AI Elite has been popping up in more conversations, and since I’ve played around with similar platforms like Bitcoin Era and Immediate Edge, I figured I’d give this one a shot and offer a breakdown that doesn’t feel like a sales pitch.

In this review, I’ll go through what Quantum AI Elite actually is, who’s behind it, what it does well, where it might fall short, and whether you should even bother signing up. Let’s dive in.

👉 Open Your Quantum AI Elite Account Now

Summary

What is Quantum AI Elite?

Quantum AI Elite is an automated trading platform that uses artificial intelligence to analyze financial markets and place trades on behalf of users. Think of it like a virtual trader that never sleeps, constantly scanning for opportunities in crypto, forex, stocks, and even commodities.

What sets it apart is the promise of using quantum computing-inspired algorithms to forecast market trends with higher accuracy. While we’re not talking real quantum computers (yet), the tech behind it is said to be faster and more adaptive than traditional algorithmic trading bots.

It’s designed for people who want to get into trading but don’t necessarily have the time (or patience) to learn candlestick patterns and economic indicators. With Quantum AI Elite, you can pretty much set your preferences, fund your account, and let the AI handle the rest.

Who Created Quantum AI Elite?

There's not a whole lot of public info about the exact developers behind Quantum AI Elite, which is often the case with these types of platforms. However, based on what I could dig up, it seems to be developed by a team of quant traders, AI engineers, and financial analysts—probably the same crowd behind earlier platforms like Bitcoin Code or Bitcoin Era.

It's partnered with regulated brokers, which is a good sign, and the platform architecture resembles that of other legit trading bots. While anonymity in the crypto space isn't unusual, I do wish they were more transparent about the dev team. Still, the performance and feedback I've seen so far speaks louder than the lack of names.

🔥 Start Trading with Quantum AI Elite Today

How Does Quantum AI Elite Work?

Quantum AI Elite works by plugging into global financial markets and scanning them in real-time using AI-driven algorithms. It looks for small but profitable patterns that most human traders would miss, and it does this across multiple asset classes.

Here’s a quick breakdown of the workflow:

• Market Analysis: The AI scans thousands of data points per second.
• Trade Execution: Once it finds a promising opportunity, it automatically places a trade.
• Risk Management: You can set stop-loss and take-profit limits to control your exposure.
• Learning Loop: The system adapts over time, learning from previous trades to improve future accuracy.

All of this happens behind the scenes, so even if you’re completely new to trading, you can still participate without being glued to charts all day.

Quantum AI Elite Pros and Cons

Pros:

• Automated trading saves time and effort
• Beginner-friendly interface
• Supports multiple asset classes
• Offers a demo account to practice before going live
• 24/7 customer support
• Positive user reviews and active community

Cons:

• Not regulated (common in crypto platforms)
• Minimum deposit of $250 might be steep for casual users
• No mobile app (mobile browser version only)
• Limited transparency about the development team

👉 Open Your Quantum AI Elite Account Now

What Devices Can be Used to Access Quantum AI Elite?

Quantum AI Elite is web-based, meaning it doesn't require any downloads or installations. You can access it from:

• Desktops (Windows, Mac)
• Tablets (iOS, Android)
• Mobile phones via web browser

The platform is fully responsive, so it adjusts well to different screen sizes. While there’s no dedicated mobile app yet, the mobile site works smoothly enough for on-the-go trading.

Quantum AI Elite – Supported Countries

Quantum AI Elite is available in most countries, including the US, UK, Canada, Australia, and across Europe. That said, due to regulatory rules, it’s not available in countries like North Korea, Iran, and a few others.

You’ll know right away during signup whether your country is supported. If not, it just blocks you from registering—no guessing games.

Quantum AI Elite – Top Features

Real-Time Market Analysis

Quantum AI Elite continuously scans the markets using real-time data pulled from multiple exchanges and news sources. This allows the AI to react quickly to price swings and news events that could impact asset value.

This feature is particularly useful for day traders or swing traders who rely on timely entries and exits.

User-Friendly Interface

The interface is clean, intuitive, and doesn’t overwhelm you with too many options. Even if you’ve never traded before, you can navigate the dashboard easily, check your performance, and adjust your settings with just a few clicks.

Mobile Accessibility

While there’s no app yet, the mobile-optimized browser version works surprisingly well. You can monitor trades, deposit or withdraw funds, and even tweak your AI settings right from your phone.

Customizable Alerts

You can set alerts for price movements, trade outcomes, and market news. These come in handy if you want to stay in the loop without constantly checking the platform.

Multiple Asset Trading

Quantum AI Elite supports trading in:

• Cryptocurrencies (Bitcoin, Ethereum, Litecoin, etc.)
• Forex pairs
• Commodities like gold and oil
• Stocks from major global markets

This diversity helps spread your risk and gives you more opportunities to profit.

Is Quantum AI Elite a Scam?

Honestly, I had that same question at first. But after testing it myself and reading through user feedback, it seems to be a legit automated trading platform. It doesn’t promise overnight wealth and offers a demo mode—two good signs.

Still, it’s important to note that it’s not government-regulated, so your funds are held by third-party brokers. Always do your own research and only deposit what you can afford to lose.

What is the Minimum Deposit Required on Quantum AI Elite?

The minimum deposit is $250, which is pretty standard for platforms in this space. This money is not a fee—it’s used as your initial trading capital.

You can deposit via:

• Credit/debit cards
• Bank transfer
• E-wallets (varies by country)

Quantum AI Elite Customer Support

Support is available 24/7 via:

• Live chat
• Email
• Callback requests

I tested their live chat during a busy market period and got a response within 2 minutes. Not bad at all.

How do you start trading on Quantum AI Elite?

Step 1: Sign Up for a Free Account

Go to the official Quantum AI Elite site, fill in your name, email, and phone number. It takes less than 2 minutes.

Step 2: Verify and Fund Your Account

You’ll be connected to a partner broker who may ask for identity verification (standard KYC process). Then deposit your $250 to get started.

Step 3: Start Trading

Once funded, you can either:

• Use the demo mode to try things out risk-free
• Activate live trading with customizable risk settings

How to Delete a Quantum AI Elite Account?

If you’re done or just want out, you can request deletion by:

• Contacting customer support
• Submitting a request through your account dashboard

They typically process deletion requests within 24–48 hours.

🔥 Start Trading with Quantum AI Elite Today

The Verdict

So, is Quantum AI Elite worth it? If you’re a beginner or even an intermediate trader looking for a semi-passive way to trade, I’d say yes.

It’s got a good mix of automation and user control, supports multiple asset types, and doesn’t drown you in technical jargon. While it’s not 100% transparent or regulated, that’s par for the course in this space.

Just remember: No AI is perfect, and you should always trade responsibly. Use the demo mode, start small, and don’t expect miracles overnight.

Frequently Asked Questions

What is Quantum AI Elite?

It’s an AI-powered trading bot that automates buying and selling of crypto, forex, and more using advanced algorithms.

How does Quantum AI Elite generate profits?

It scans markets for price patterns and executes trades automatically using predefined strategies and risk settings.

Is Quantum AI Elite safe to use?

Yes, as long as you trade responsibly. The platform uses SSL encryption and works with regulated brokers.

What is the minimum investment required to start with Quantum AI Elite?

You’ll need to deposit $250 to activate live trading.

Can I use Quantum AI Elite on my mobile device?

Yep! It's fully web-based and works great on mobile browsers.

How do I contact customer support for Quantum AI Elite?

You can reach them via live chat, email, or by requesting a callback.

Are there any hidden fees with Quantum AI Elite?

There are no hidden fees. Some partner brokers may charge small commissions, but the platform itself is free to use.

What trading assets can I access with Quantum AI Elite?

You can trade crypto, forex, stocks, and commodities.

How do I withdraw funds from my Quantum AI Elite account?

Just go to your dashboard, click on Withdraw, and follow the steps. Processing usually takes 24–48 hours.

Can beginners use Quantum AI Elite effectively?

Absolutely. It’s designed to be beginner-friendly, with a demo mode and simplified controls.

Top Comment: stop dude

Forum: r/dallasmavericks