Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

I used AI-powered calorie counting apps, and they were even worse than expected

lifehacker.com

Post reply on HN
Print view View original

291–300 of 302 posts

Re: I used AI-powered calorie counting apps, and they were even worse than expected

#291

Post by n4r9 »

Kirby64 wrote: post #288
n4r9 wrote: post #284

Earlier quoted context omitted.

That's nonsense. A shot of spirits is one unit of alcohol. A pint of strong-ish beer like Stella Artois is 3 units. 550 calories is easily a dinner's worth for a woman of 70kg, or a man that's heavier but trying to gradually lose weight. Either of whom would normally not be too affected by two pints of Stella.

> A pint of strong-ish beer like Stella Artois is 3 units. No it's not. Stella is 5.2% in the strongest form (some places sell a weaker version that is 4.6%). Both are roughly standard ABVs. At worst, a 20 oz 'pint' of Stella would be ~1.7 units. In most places in the US, a pint is 16 oz, so it would be ~1.4 units. Two pints at worst is slightly over 3 units total.

Maybe I'm misreading it (or it's just wrong) but this NHS webpage very clearly states that a "Pint of higher-strength lager/beer/cider (ABV 5.2%)" has 3 units:

https://www.nhs.uk/live-well/alcohol-advice/calculating-alco...

Re: I used AI-powered calorie counting apps, and they were even worse than expected

#292

Post by Kirby64 »

n4r9 wrote: post #291
Kirby64 wrote: post #288

Earlier quoted context omitted.

> A pint of strong-ish beer like Stella Artois is 3 units. No it's not. Stella is 5.2% in the strongest form (some places sell a weaker version that is 4.6%). Both are roughly standard ABVs. At worst, a 20 oz 'pint' of Stella would be ~1.7 units. In most places in the US, a pint is 16 oz, so it would be ~1.4 units. Two pints at worst is slightly over 3 units total.

Maybe I'm misreading it (or it's just wrong) but this NHS webpage very clearly states that a "Pint of higher-strength lager/beer/cider (ABV 5.2%)" has 3 units: https://www.nhs.uk/live-well/alcohol-advice/calculating-alco...

Ah, right. We're both getting it a bit wrong. A 'unit' (10 ml of pure ethanol) isn't actually equivalent to a 'drink'. One 'drink' is considered to be 12 oz of 5% beer, 5 oz of 12% wine, or one 1.5 oz 40% shot. None of those are actually '1 unit', they're all ~1.7 units (~17 ml of ethanol).

I think most folks think in 'drinks' and not 'units'.

So, back to the original comment: 6 shots at most establishments would be ~10 'units' of alcohol. Or, closer to 3.3 20oz pints of beer. 3.3 20oz beers would be enough to get most folks pretty intoxicated if drank quickly; the difference being that 3.3 20oz beers (66oz total!) is a lot harder to drink very quickly than 6 shots, purely by volume of liquid.

Re: I used AI-powered calorie counting apps, and they were even worse than expected

#293

Post by n4r9 »

Kirby64 wrote: post #292
n4r9 wrote: post #291

Earlier quoted context omitted.

Maybe I'm misreading it (or it's just wrong) but this NHS webpage very clearly states that a "Pint of higher-strength lager/beer/cider (ABV 5.2%)" has 3 units: https://www.nhs.uk/live-well/alcohol-advice/calculating-alco...

Ah, right. We're both getting it a bit wrong. A 'unit' (10 ml of pure ethanol) isn't actually equivalent to a 'drink'. One 'drink' is considered to be 12 oz of 5% beer, 5 oz of 12% wine, or one 1.5 oz 40% shot. None of those are actually '1 unit', they're all ~1.7 units (~17 ml of ethanol). I think most folks think in 'drinks' and not 'units'. So, back to the original comment: 6 shots at most establishments would be…

Thanks for the clarification. Though on a lighter personal note I'd definitely struggle to drink 6 shots of gin (or tequila for that matter) in the same day. Awful stuff!

Re: I used AI-powered calorie counting apps, and they were even worse than expected

#294

Post by shifto »

wnorris510 wrote: post #272
JohnMakin wrote: post #223

Earlier quoted context omitted.

This is the classic refrain product owners counter back with any time people present reasonable criticism of their AI app. User: "This (AI product) doesn't work!" Product Owner: "Well, humans are also bad at that." That's not the promise of these apps in general! The whole selling point of AI is that they're vastly better - if my eyeball estimate is "pretty inaccurate" and by your own admission the app is "pretty ina…

I got into this space because I saw many family members struggling with weight and nutrition. One of the leading causes of death in the United States is Diabetes, largely caused by poor nutrition and lack of exercise. I've spent the last 10 years of my life researching why this is and how I can help. What I've heard from countless users is that it takes to long to track what they're eating and learn how they can impr…

What is the added value of the app? Ive been using ChatGPT to track my calories for 6 months now and dropped 20kg because of it. This is mostly me just telling it everything that enters my mouth and adjust the output I deem wrong. This isn't super accurate but apparently enough to get an idea, and thus lose weight.

Re: I used AI-powered calorie counting apps, and they were even worse than expected

#295

Post by wnorris510 »

shifto wrote: post #294
wnorris510 wrote: post #272

Earlier quoted context omitted.

I got into this space because I saw many family members struggling with weight and nutrition. One of the leading causes of death in the United States is Diabetes, largely caused by poor nutrition and lack of exercise. I've spent the last 10 years of my life researching why this is and how I can help. What I've heard from countless users is that it takes to long to track what they're eating and learn how they can impr…

What is the added value of the app? Ive been using ChatGPT to track my calories for 6 months now and dropped 20kg because of it. This is mostly me just telling it everything that enters my mouth and adjust the output I deem wrong. This isn't super accurate but apparently enough to get an idea, and thus lose weight.

Our app uses LiDAR for portion and is a lot more accurate than ChatGPT, ChatGPT doesn't have access to LiDAR and is less accurate.

Our app is free to use the most accurate model, ChatGPT is paid for the most accurate model and has usage limits.

Our app plots your data on nice charts and has visuals specific to understanding your data over time visually (e.g. how has my sodium intake been changing day over day or what has my average weekly caloric intake been when deducting active calories from fitness trackers).

Our app integrates with Apple Health and soon will integrate with Android Fit so you can export your data to other apps, share with health professionals, etc.

Re: I used AI-powered calorie counting apps, and they were even worse than expected

#296

Post by wnorris510 »

sumeno wrote: post #225
wnorris510 wrote: post #194

Earlier quoted context omitted.

We do a lot of research on preventing eating disorders. As the article mentions we did not suggest she lose weight down to being underweight like other apps did. Our app has the fruit and veg counters ABOVE the calorie meter on the dashboard. The reason for this is that maximization mindset (e.g. maximizing fruits and veg) is way healthier than a minimization mindset (e.g. minimizing calories or carbs). We actually e…

You are still selling a product that says it can count calories when fundamentally it cannot. The fact that people believe you and pay you money for it doesn't change that fact. You are lying to people. > SnapCalorie is the first app where you can take a picture of any meal and get an accurate calorie count and nutrition in seconds This is from your website. It is pure fiction. You admit as much in your first post

We published in the top conference of computer vision a peer reviewed study of our accuracy. On average we are twice the accuracy of a professional nutritionist. The product is free to try. If you're not happy with the accuracy, don't pay! We have around half a million people each month coming to our app and a vast majority stick around and are happy with the results.

If there's something we missed on your food shoot me a DM, I'd love to dig in.

Re: I used AI-powered calorie counting apps, and they were even worse than expected

#297

Post by wnorris510 »

mgraczyk wrote: post #276
wnorris510 wrote: post #275

Earlier quoted context omitted.

You have access to all of the same features without upgrading to premium. Just click the "x" on the paywall.

The paywall says I get extra stuff from premium, including "unlimited photo, voice, and barcode logging", "unlimited AI nutritionist chat", "unlimited Personalized tips from your AI nutritionist". It sounds like if I pay more, I get more access and can use the app more? So the original article was completely correct in characterizing the premium plan as $79.99 It says > The app was free to download, no trial period n…

Apple requires us to say you get more as part of premium. They have silly rules that tie our hands on stuff. You have access to everything as part of the free tier. Hopefully the recent court case from Riot will help make it easier for us to clarify this.

Re: I used AI-powered calorie counting apps, and they were even worse than expected

#298

Post by mgraczyk »

wnorris510 wrote: post #297
mgraczyk wrote: post #276

Earlier quoted context omitted.

The paywall says I get extra stuff from premium, including "unlimited photo, voice, and barcode logging", "unlimited AI nutritionist chat", "unlimited Personalized tips from your AI nutritionist". It sounds like if I pay more, I get more access and can use the app more? So the original article was completely correct in characterizing the premium plan as $79.99 It says > The app was free to download, no trial period n…

Apple requires us to say you get more as part of premium. They have silly rules that tie our hands on stuff. You have access to everything as part of the free tier. Hopefully the recent court case from Riot will help make it easier for us to clarify this.

I just tried in the app on Android and it would not allow me to take more photos unless I pay.

When you say "you have access to everything as part of the free tier", does that apply to Android as well?

Because what you are saying is not true on Android. After 3 photos I cannot take another, the message says "You used all of your photo logs! As a free tier member you get 3 free Photo logs per day."

You are claiming that this is different on iOS?

Re: I used AI-powered calorie counting apps, and they were even worse than expected

#300

Post by boydYVR »

mkfs wrote: post #89

Fat is the most calorie-dense macro at 9 per gram (protein and carb are both 4), so your accuracy is going to be heavily skewed by how accurately you can identify the fat content of the meal. The problem is that it's often in forms like oils that don't seem well-suited to AI image recognition.

Yeah. Relying on images of food to track calories is like ... dancing about architecture.
Post reply on HN