Are you getting advice from AI on Reefing questions?

[Randy Holmes-Farley]

Die selbe Berechnung in chatgGPT. Ist correct?

1. Mass of Water in a 10-Gallon Reef Aquarium

1 gallon of water = 3.785 liters

10 gallons = 37.85 liters

Density of seawater (PSU 35) = 1.025 g/ml = 1.025 kg/L

Mass = 37.85 liters × 1.025 kg/L = 38.8 kg

2. Mass of Air in a 12 x 12 x 8 ft Room

Room volume:

12 ft × 12 ft × 8 ft = 1,152 ft³

1 cubic foot = 28.3 liters

1,152 ft³ × 28.3 L/ft³ = 32,582 liters


Density of air at 25°C = 1.2 g/L = 0.0012 kg/L

Mass of air = 32,582 L × 1.2 g/L / 1000 = 39.12 kg

Not sure exactly what you asked it, but asking about water and giving a mass answer that includes salts is a partial fail to what I asked in the question of the day, IMO.

 

[Luminous74]

Not sure exactly what you asked it, but asking about water and giving a mass answer that includes salts is a partial fail to what I asked in the question of the day, IMO.

Then I was the one who failed, not the KI, since I assumed it was seawater.

 

[Randy Holmes-Farley]

Then I was the one who failed, not the KI, since I assumed it was seawater.

The material in the tank in the question I asked (see below) is presumably seawater. The question was about "water", not other stuff that might be in the water.

What did you literally ask the AI?

Here's the question I asked:

Which has more mass:

1. The water in a 10 gallon reef aquarium
2. The air in a 12 ft x 12 ft x 8 foot (l x w x h) room?

and that is not the same question as:

Which has more mass:

1. The seawater in a 10 gallon reef aquarium
2. The air in a 12 ft x 12 ft x 8 foot (l x w x h) room?

 

[Luminous74]

The material in the tank in the question I asked (see below) is presumably seawater. The question was about "water", not other stuff that might be in the water.

What did you literally ask the AI?

Here's the question I asked:

Which has more mass:

1. The water in a 10 gallon reef aquarium
2. The air in a 12 ft x 12 ft x 8 foot (l x w x h) room?

and that is not the same question as:

Which has more mass:

1. The seawater in a 10 gallon reef aquarium
2. The air in a 12 ft x 12 ft x 8 foot (l x w x h) room?

That's exactly what I asked, just as they formulated it. But you're right, it's not the same question. You remind me of my math teacher — meant with humor. Thank you!

------------------------------------------------------------------------------------------
Random Reef Chemistry Question of the Day

Which has more mass:

1. The water in a 10 gallon reef aquarium
2. The air in a 12 ft x 12 ft x 8 foot (l x w x h) room?
--------------------------------------------------------------------------------------------

 

[Randy Holmes-Farley]

Thanks.

 

[Cool tangs]

AI is a interesting topic. Arguably true AI would be able to formulate its own decisions. I'd argue things like copilot, chat gpt, etc are more like machine learning then AI. It needs to be fed data to produce data and the data it produces is only as good as the data supplied.

Now as for do I think it's morally right to sell marketed AI products that regurgitate other people's work no. I view it more as a tool to save you the work load of searching hours on end when it can give (some days), okay answers or a basis to work off of. Most of the time I find things like chatgpt and copilot need a lot of coercion and corrections. I use it a lot in my field of work, but it's by means not going to help someone who doesn't understand what they are asking. In my opinion anyway.

Now on the other side if the product is free and it's using others information. Well you already consented the moment you used the WWW(world wide web) to post any data in a forum, a website, chat, social media etc, etc. It's open for everyone and anyone to see. Just like when you sign up for your services you've probably signed a EULA (end user license agreements) without reading the fine print and agreed to give up your personal data for marketing and on selling. Meaning you don't own it.

Sadly the WWW is not how it was intended to be used when it was first invented. But as a online gamer and a cautious consumer who tries to avoid social media, I also can't live without it

 

[Dragen Fiend]

I just google it. Now majority of the time I go through articles or sites discussing the topic I'm searching. However, Google uses CHATGPT at the top for basic AI answers. What I do is read it. Then I further research it to confirm its validity.

 

[drewzaun]

A thought on science:
I’m not a scientist, so I can be a little cynical at times. I don’t blindly trust science. As the saying goes, to err is human. Science has been wrong many times before—even in its calculations.

I personally see AI as a valuable tool for me, and I’m excited to see how it continues to develop.

I’m sure AI has value in certain situations outside my experience, and I can see that it will almost certainly never go away and will become more usable every day. I also don’t think everything being packaged as “AI” actually is.

As for science, science is not a statement but a process to learn about the natural world we live in. It’s the best tool we have to answer any relevant questions and the process is mainly a guy making a statement and everyone else trying to make him look stupid (I know, I know, but it kind of is really)

What most people argue is a flaw in science, its ability to pivot on the discovery of new information, is its best value. Every bit of data must be accounted for. Where it goes wrong is when it is being misused by corporations, politicians, the media and people who just plain don’t want to believe it, for an agenda. For example take climate change. That our weather is different than it was a short few decades ago is an unmistakable observation. For some reason I don’t understand it has become a partisan political debate, with one side acting like an ostrich and the other side acting like a doomsday cult leader. Neither argument is correct, no scientists really agree with any of them, there are bits of reality in both arguments but none of it is science despite the insistence of both parties. While the esteemed idiots in DC use it as political fodder the rest of us have to deal with the consequences of all these “100 year events” that we’ve had in the last 20 years and not see any help from the only people that have the ability to really change anything. Because IMO the “debate” is more valuable than the reality we see daily. Not to get political. I hate both parties so there.

Like a hammer, science is a useful and valuable tool, but it can be abused and misused. What you don’t trust is probably not the science but the way it was presented to you, by a non scientist.

Don’t kill me lol

 

[Randy Holmes-Farley]

As for science, science is not a statement but a process to learn about the natural world we live in.

I agree. Folks misunderstand science and what it does and does not do. We are getting off topic here, but I think this is a big issue in reefing, where folks confuse the veracity of observations (which are only wrong if they are made up) with the veracity of conclusions that one might draw from them (which are often wrong).

Observed Effects are Real; Explanations May Not Be

This has come up time and again in all sorts of reefing discussions, from manufacturer claims to hobbyist discussions of their systems. 1. Anything someone observes is real, and has some real explanation (assuming its not a lie) 2. Any explanation of an observation should ALWAYS be treated...

www.reef2reef.com

 

[MagicPenny]

I have used it for setup questions, and I found it to be pretty useful, but I think we are all also looking at it with a) a healthy dose of skepticism, b) at least some knowledge of our own, and c) knowing that AI is not the gospel on anything. That is key - vetting the info you receive.
My question, for instance, was how to keep my RO/DI from freezing up in my detached garage. I had originally planned to get a clamp-on light and a heat bulb and point it at the unit. AI suggested a seedling mat wrapped around it, which uses less power and would be safer since it's waterproof. Prior to that I didn't know what a seedling mat was, but now I own one.
Then it asked me what fish I was planning on, and gave intelligent answers when I prodded it about species I already knew the answer about. It's a little weird knowing you're talking to a computer, but the new version is pretty conversational!

 

[ogbounceman]

following

 

[whybenormal]

I often ask Gemini (Google's AI) especially when my searches for an answer just don't give me any satisfactory answers. Sometimes it is wording, even on here, so I have a little chat with Gemini, and get some answers, then I usually have a better way to phrase a question to find more of what I want.

A few weeks ago I was having trouble getting my brine shrimp to grow to adult. I kept googling, getting nowhere. I asked Gemini, got a list of steps, and after everything it seems my issue was temperature.
I probably use AI a good bit in research - much like the rest of you, a glorified search engine - but I'm at my desk a lot in my other life as a business coach.

 

[Luminous74]

I’m sure AI has value in certain situations outside my experience, and I can see that it will almost certainly never go away and will become more usable every day. I also don’t think everything being packaged as “AI” actually is.

As for science, science is not a statement but a process to learn about the natural world we live in. It’s the best tool we have to answer any relevant questions and the process is mainly a guy making a statement and everyone else trying to make him look stupid (I know, I know, but it kind of is really)

What most people argue is a flaw in science, its ability to pivot on the discovery of new information, is its best value. Every bit of data must be accounted for. Where it goes wrong is when it is being misused by corporations, politicians, the media and people who just plain don’t want to believe it, for an agenda. For example take climate change. That our weather is different than it was a short few decades ago is an unmistakable observation. For some reason I don’t understand it has become a partisan political debate, with one side acting like an ostrich and the other side acting like a doomsday cult leader. Neither argument is correct, no scientists really agree with any of them, there are bits of reality in both arguments but none of it is science despite the insistence of both parties. While the esteemed idiots in DC use it as political fodder the rest of us have to deal with the consequences of all these “100 year events” that we’ve had in the last 20 years and not see any help from the only people that have the ability to really change anything. Because IMO the “debate” is more valuable than the reality we see daily. Not to get political. I hate both parties so there.

Like a hammer, science is a useful and valuable tool, but it can be abused and misused. What you don’t trust is probably not the science but the way it was presented to you, by a non scientist.

Don’t kill me lol

I never try to be upset with anyone; instead, I try to understand them.


Thank you very much for your contribution — you’ve spoken from my heart on many points, even though we come from quite different parts of the world

 

[GARRIGA]

AI consolidates answers based on various sources without knowing which are valid sources and although I use it to see what's driving the stock market it does provide links to those sources I can then validate for accuracy and use it as a means of obtaining an executive summary with zero confidence the information being accurate.

 

[Luminous74]

.

 

[hsp]

Sure. I do not know which versions of ChatGPT were used, but it has made some terrible math mistakes in answering reef dosing questions that I had to correct, and I have a thread with the questions I asked Chat GPT and I graded its answers.

Here's a fun example from the first link below:

1. Does saturated kalkwasser have a higher pH than saturated sodium hydroxide?

ChatGPT reply:

Saturated kalkwasser (calcium hydroxide) has a higher pH than saturated sodium hydroxide (NaOH). The pH of saturated kalkwasser is around 12.4, while the pH of saturated sodium hydroxide is around 14.

Randy Rating: 0% fail. It doesn't know 14 is higher than 12.4? 14 is also wrong. NaOH saturation pH is higher than 14.

Randy's Questions ChatGPT and grades the answers

ChatGPT and Reef Chemistry: We Try it Out!

Why didn't I think of this sooner. No need to have a real person helping out. lol Here are a few examples that I asked it today, in order of increasing difficulty (IMO): 1. Describe the importance of alkalinity and calcium in a coral reef aquarium. ChatGPT reply: Alkalinity and calcium are...

www.reef2reef.com

Math Mistake, dosing off by 27,000x

Potassium chloride to raise potassium

If I add 2g of potassium chloride into my 260 gallon tank, how many ppm will potassium increase? I asked ChatGPT, and it said this. “ Thank you for clarifying. Let's calculate the increase in potassium concentration when adding 2 grams of potassium chloride (KCl) to 984 liters: The molar mass...

www.reef2reef.com

Thanks, Randy! I see this post is from 2023. Humans had the last hardware upgrade 160,000 years ago. AI gets one every 2 months or so. So I run the question again:

==========
Does saturated kalkwasser have a higher pH than saturated sodium hydroxide?
---------
ChatGPT4o:
Yes, saturated sodium hydroxide (NaOH) solution has a significantly higher pH than saturated kalkwasser (calcium hydroxide, Ca(OH)₂).

Here’s why:
1. pH of Saturated Kalkwasser (Calcium Hydroxide)
• Ca(OH)₂ has a solubility of ~1.73 g/L in water at 25°C, leading to a saturated solution with a pH of around 12.4 to 12.6.
• The pH is limited by the equilibrium between Ca(OH)₂ and its dissociation into Ca²⁺ and OH⁻ ions.

2. pH of Saturated Sodium Hydroxide
• NaOH is far more soluble (~1,110 g/L at 25°C) and completely dissociates into Na⁺ and OH⁻ ions, producing an extremely alkaline solution.
• A 50% NaOH solution (approximately 19M NaOH) has a pH close to 14 to 15.
• Even a 1M NaOH solution has a pH of about 14.

Comparison
• Saturated NaOH (~50%) → pH ~14-15
• Saturated Ca(OH)₂ (kalkwasser) → pH ~12.4-12.6

Thus, sodium hydroxide produces a much higher pH than kalkwasser. Kalkwasser is often used in reef aquariums for maintaining alkalinity and calcium levels, but its pH remains lower due to its much lower solubility compared to NaOH.
=======

Your judgement?

 

[Randy Holmes-Farley]

Your judgement?

The new version probably read my abuse of the old version and learned how to answer that question. lol

 

[hsp]

The new version probably read my abuse of the old version and learned how to answer that question. lol

Haha! It takes a few months to develop these models. So clearly not.

 

[Randy Holmes-Farley]

Haha! It takes a few months to develop these models. So clearly not.

My criticism of its answer to that question was, as you said, in 2023, so a later version could literally have read my criticism, but it also may just be a better chemist than earlier versions.

 

[Opus]

There's a dryness to it. That sedated, blunted affect seemingly created by changing the page breaks and bullet point headers + background art of whole articles returned by chatgpt

He he

Really though, if that's a real person in those articles I apologizeish. This is merely the gauntlet expected, total sarcastic rejection until I get shown that isn't ai. It seems like ai 1000000000000000000%

All writers need to be able to reef on command, or retire. By reef on command am meaning in addition to writing, and posting, I see you alongside us in work threads taking accountability for outcomes you predicted in your published articles.

If someone writes about how to fix gha and i can't find ten threads of them fixing it for others: they're getting ribbed mercilessly

The person that puts up those articles is a marine scientist. She is also a curator of the 75,000 gallon IBOC aquarium in Dallas. Mantra is her company name which seems to be some kind of marine digital creation company. I wouldn't be surprised if she used AI to create the articles. Not sure how much she edits them if they are AI created. I've never actually read one. She does do live talks at reef clubs and I think she did a talk at either RAP or Aquashella last year. So I think she has the qualifications, but again she could very well be using just AI for the articles.
I do agree, AI articles do usually lack any kind of "personality" to them.