Pls get rid of the “critical thinking app (dot) ages 10-14” thing at the top of the hero. It’s one of the most obvious signs a site was generated by an LLM. It also seems like the text on the site (and your post here) is AI-generated, which doesn’t inspire confidence.
I’m a fan of the idea behind this, though. Cool project.
I've come around. I think there's really nothing wrong with guiding AI to develop a speculative, experimental, personal project. You're going to see rough edges, but guess what? You see rough edges when you write your own code, too. The process of fixing them is the same process either way: see and adjust.
You are in a violation of GDPR because you don't show "allow all" and "decline all" buttons equally. I won't look up the exact regulation quote, but it is somewhere there.
Also, why do you need to store personal information of the users in the first place? A SAAS operator can handle user payments just fine without any extra data stored to require an additional consent, because storing payment status doesn't require it by law.
We do not gather payment information like card numbers etc, but we collect events from Stripe Payments if the payment was successful to enable the respective tier. why do you say that?
Reading an AI generated text and discovering that is generated is a skill. A skill that the pedagogical system from any country should promote. Lathoa does not teach, but it is a fun way to work out your critical thinking skill.
> Pls get rid of the “critical thinking app (dot) ages 10-14” thing at the top of the hero. It’s one of the most obvious signs a site was generated by an LLM.
If you’ve made something with an LLM, it should be obvious you did (preferably with an outright admission, though, not by leaving artefacts for people to deduce from) and not try to hide it.
Either you believe building with LLMs is OK or you don’t. If you do think it’s OK, then own up to it. If you don’t think it’s OK, don’t do it.
Reminds me of LIGO, which injected fake results until the humans got good at handling them, then later announced that one of the results wasn't fake. Seems at first like an upside-down way to discover gravitational waves, but I think it's a pattern worth exploring in more cases.
To paraphrase Doctorow:
> This is why the TSA are some of the most water-bottle-findingest **ers, but regularly overlook firearms when a red team takes one through. Humans do not stay vigilant to signals which almost never happen.
This is why not all cases are wrong. Too few of those and kids will learn to always claim an error; too many and they will stop looking. There is a balance..
This should be good training. Because the kids will encounter many, many more LLMs that are wrong on purpose, for the sake of propaganda and fraud. Teaching them early on that they can’t trust the machines is a net positive.
I share the same viewpoint. Thank you. The education systems will change, and have been constantly changing. Lathoa tries to keep an equilibrium here, or at least keep something alive..our brain..
I think speed would be irrelevant, or perhaps would discourage deep engagement with the problems.
I think presenting it as AI can be wrong is not as valuable as presenting it as developing the ability to question what one is told. (e.g. Verizon Math)
indeed I agree with you however studies have shown that kids have access to interfaces to chat with an LLM from a young age, and the traction this technology is taking is increasing. Indeed verizon math seems more whole when it comes to topic variety.
This is an interesting idea, but I need more than one sample to get a feel for it. I did the example on the homepage, and then one more that it linked to, and both had extremely basic errors my kid would have no problem with (currently doing Math Academy prealgebra, approx year 9/8th grade level). I'd suggest allowing people to try more than one very basic error.
Thanks! You may sign up and get on the free plan for three cases per day (only the obvious level). You can subscribe to any plan and get a 7 day free trial of that plan also. cancel any time within the trial period.
Reminds me of https://en.wikipedia.org/wiki/QAMA_Calculator - a calculator that won't reveal the precise answer until you supply an estimate (physical hardware in 2014, I think the current version is an app.) The website makes claims about the pedagogic value, but I didn't see any actual citations at https://qama.world/the-science/ just "everyone knows" kind of things... but if they had research supporting it that would probably inform your approach?
I like the idea. But it’s wrong on today problem (movie started at 3:45 and lasts 90 min. When it will end?) - first choice 3:45 + 90 min was rejected by the program! The two other choices are even more problematic!
And I think that highlights the problem with AI tutor: no definite answer because no epistemic understanding. And that can undermine the trust in the system.
On the full problem solving screen, the little robot mascot bounces up and down. However, the shadow moves along with the robot. It looks a bit odd and I think it might work better anyways if the shadow stays in place and shrinks/grows according to the robots bouncing.Other than that, seems really well put together.
I did not understand "today's case" right away. I thought you were supposed to answer which of the three cards was wrong. Didn't realize it was showing a series of steps the student worked through. As someone else pointed out, khan does have questions like this
yes yes these three cards are steps of one one solution, top to bottom; similarly to a way of thought. Some may sound unnecessary or silly, but that is up to the kid's judgement. Khan has these too, but Lathoa has a different presentation style.
while solving a lathoa case, if you spot the error you get back an amount of points. if you score also the explanation exhibiting your understanding you gain more points.
XP moves a kid up the detective ranks, from Rookie to Sherlock. At 2000XP a leaderboard unlocks, but only if a parent turns it on. Nothing to buy there..
Not all cases are wrong though. One in five is correct and the "Looks correct" scores also and gets XP. So it does practise knowing when not to doubt..a little..
Pls get rid of the “critical thinking app (dot) ages 10-14” thing at the top of the hero. It’s one of the most obvious signs a site was generated by an LLM. It also seems like the text on the site (and your post here) is AI-generated, which doesn’t inspire confidence.
I’m a fan of the idea behind this, though. Cool project.
I've come around. I think there's really nothing wrong with guiding AI to develop a speculative, experimental, personal project. You're going to see rough edges, but guess what? You see rough edges when you write your own code, too. The process of fixing them is the same process either way: see and adjust.
Not exactly a personal project, though. This is a SaaS with a registered trademark and multiple paid plans, up to $150 per year.
That's orthogonal. Anyone can make a LLC, file a patent, apply for a trademark, do it any way you want.
it is a paid product indeed, but you can try it via the free plan forever. Also new accounts get 7 days of the Parent plan with no card.
glad that you like the idea! Indeed Lathoa is built with a lot of help from Claude. Thanks for the suggestion regarding the pill on the homepage.
You are in a violation of GDPR because you don't show "allow all" and "decline all" buttons equally. I won't look up the exact regulation quote, but it is somewhere there.
Also, why do you need to store personal information of the users in the first place? A SAAS operator can handle user payments just fine without any extra data stored to require an additional consent, because storing payment status doesn't require it by law.
Wow great catch! Indeed! Thanks for spotting that out! I will fix it promptly!
Regarding the cookies there is a full description in the privacy policy page here: https://lathoa.ai/en/privacy
We do not gather payment information like card numbers etc, but we collect events from Stripe Payments if the payment was successful to enable the respective tier. why do you say that?
To be fair the entire site screams Claude
Yes, it may help kids learn math but it also teaches them to speak like a LLM.
Reading an AI generated text and discovering that is generated is a skill. A skill that the pedagogical system from any country should promote. Lathoa does not teach, but it is a fun way to work out your critical thinking skill.
Yes that is fair. It has been built with Claude's assistance.
> Pls get rid of the “critical thinking app (dot) ages 10-14” thing at the top of the hero. It’s one of the most obvious signs a site was generated by an LLM.
If you’ve made something with an LLM, it should be obvious you did (preferably with an outright admission, though, not by leaving artefacts for people to deduce from) and not try to hide it.
Either you believe building with LLMs is OK or you don’t. If you do think it’s OK, then own up to it. If you don’t think it’s OK, don’t do it.
Reminds me of LIGO, which injected fake results until the humans got good at handling them, then later announced that one of the results wasn't fake. Seems at first like an upside-down way to discover gravitational waves, but I think it's a pattern worth exploring in more cases.
To paraphrase Doctorow:
> This is why the TSA are some of the most water-bottle-findingest **ers, but regularly overlook firearms when a red team takes one through. Humans do not stay vigilant to signals which almost never happen.
This is why not all cases are wrong. Too few of those and kids will learn to always claim an error; too many and they will stop looking. There is a balance..
This should be good training. Because the kids will encounter many, many more LLMs that are wrong on purpose, for the sake of propaganda and fraud. Teaching them early on that they can’t trust the machines is a net positive.
I share the same viewpoint. Thank you. The education systems will change, and have been constantly changing. Lathoa tries to keep an equilibrium here, or at least keep something alive..our brain..
I think speed would be irrelevant, or perhaps would discourage deep engagement with the problems.
I think presenting it as AI can be wrong is not as valuable as presenting it as developing the ability to question what one is told. (e.g. Verizon Math)
indeed I agree with you however studies have shown that kids have access to interfaces to chat with an LLM from a young age, and the traction this technology is taking is increasing. Indeed verizon math seems more whole when it comes to topic variety.
regarding the speed comment, yes it is removed because eventually it would teach rushing. Thanks!
This is a category of exercise that is used in Khan Academy and in classroom math, yes.
https://learningfocused.com/blogs/lesson-planning/unlocking-...
Yes it is error analysis. However Errol critically responds back to the user's explanation for the error, with a helpful learning tone.
This is an interesting idea, but I need more than one sample to get a feel for it. I did the example on the homepage, and then one more that it linked to, and both had extremely basic errors my kid would have no problem with (currently doing Math Academy prealgebra, approx year 9/8th grade level). I'd suggest allowing people to try more than one very basic error.
Thanks! You may sign up and get on the free plan for three cases per day (only the obvious level). You can subscribe to any plan and get a 7 day free trial of that plan also. cancel any time within the trial period.
The idea is interesting, but generally, I'd still prefer that children don't use too many electronic devices. The real physical world is better..
Reminds me of https://en.wikipedia.org/wiki/QAMA_Calculator - a calculator that won't reveal the precise answer until you supply an estimate (physical hardware in 2014, I think the current version is an app.) The website makes claims about the pedagogic value, but I didn't see any actual citations at https://qama.world/the-science/ just "everyone knows" kind of things... but if they had research supporting it that would probably inform your approach?
I will read this thoroughly thanks!
I like the idea. But it’s wrong on today problem (movie started at 3:45 and lasts 90 min. When it will end?) - first choice 3:45 + 90 min was rejected by the program! The two other choices are even more problematic!
And I think that highlights the problem with AI tutor: no definite answer because no epistemic understanding. And that can undermine the trust in the system.
The steps are Errol's chain of thought. The user is called to pick the first wrong statement that leads to a wrong result.
> And that can undermine the trust in the system.
That's not a bug, it’s a feature. Trust in the system would be severely misplaced.
how so? can you please provide an example?
On the full problem solving screen, the little robot mascot bounces up and down. However, the shadow moves along with the robot. It looks a bit odd and I think it might work better anyways if the shadow stays in place and shrinks/grows according to the robots bouncing.Other than that, seems really well put together.
good catch! Thanks
I did not understand "today's case" right away. I thought you were supposed to answer which of the three cards was wrong. Didn't realize it was showing a series of steps the student worked through. As someone else pointed out, khan does have questions like this
yes yes these three cards are steps of one one solution, top to bottom; similarly to a way of thought. Some may sound unnecessary or silly, but that is up to the kid's judgement. Khan has these too, but Lathoa has a different presentation style.
speaking of LLMisms "That's the skill"
> if she finds an error to gain more XP
What does this mean?
while solving a lathoa case, if you spot the error you get back an amount of points. if you score also the explanation exhibiting your understanding you gain more points.
But what are the points used for?
Getting kids addicted to something useful.
XP moves a kid up the detective ranks, from Rookie to Sherlock. At 2000XP a leaderboard unlocks, but only if a parent turns it on. Nothing to buy there..
It helps us practice critical thinking. What it doesn’t do is help us practice knowing WHEN to use critical thinking.
Easy, all the damn time.
https://xkcd.com/309/
kek..
interesting perspective...when and how much also is a skill someone should have.
Not all cases are wrong though. One in five is correct and the "Looks correct" scores also and gets XP. So it does practise knowing when not to doubt..a little..