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

221–230 of 302 posts

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

#221

Post by sumeno »

gadders wrote: post #197
sumeno wrote: post #184

Earlier quoted context omitted.

What about when people give up because you massively underestimate their calories? Or when people develop eating disorders because you massively overestimate their calories? If the goal is to learn then accurate information is important, although I suppose it's harder to get a VC to fund.

The calorie counts on food packets themselves are +/- 20%, so it's inaccurate all the way down. Some of the value also just comes from writing down what you eat, and noting the snacks, dressings, glasses of wine etc that you forget about but all add up.

+/- 20% is way better than being off by 3x like in the example in this article

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

#222

Post by washmyelbows »

hedora wrote: post #145
SketchySeaBeast wrote: post #119

Earlier quoted context omitted.

I think when people are mentioning alcohol they typically mean beer or wine - a can of beer is ~150 calories, so have a few of those a night, which isn't at all uncommon in some households, and over a week you're up ~1 lbs.

Also, 150 calories is somewhere between cheap American beer and a nice IPA or stout. You can easily hit 200-300 calories per beer.

my favorite stout is 400 calories for 12oz

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

#223

Post by JohnMakin »

wnorris510 wrote: post #183

Founder of https://snapcalorie.com and ex-Google AI researcher here. The article actually gets our pricing wrong, it's completely free to use SnapCalorie, the $79.99 price is a donation to support our research and help us continue providing the best tools to everyone for free. We've published in top peer reviewed academic conferences and our algorithm (can't say the same for other apps) IS actually more accurate than…

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 inaccurate" then why the hell would I use your app??

From the very top of the page you linked:

> SnapCalorie is the first app where you can take a picture of any meal and get an accurate calorie count and nutrition in seconds

(emphasis mine)

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

#224

Post by ericmcer »

wnorris510 wrote: post #183

Founder of https://snapcalorie.com and ex-Google AI researcher here. The article actually gets our pricing wrong, it's completely free to use SnapCalorie, the $79.99 price is a donation to support our research and help us continue providing the best tools to everyone for free. We've published in top peer reviewed academic conferences and our algorithm (can't say the same for other apps) IS actually more accurate than…

Thanks for mentioning voice, I prefer that to photos. Photos seems like the most complex technically.

AI seems to be going heavy into text/chat and images but voice is really the perfect UI to me. I would be totally fine if I could trigger a "call" from the app and talk to an AI on the phone for 30s and it could log my meal, give me some advice, etc.

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

#225

Post by sumeno »

wnorris510 wrote: post #194
sumeno wrote: post #184

Earlier quoted context omitted.

What about when people give up because you massively underestimate their calories? Or when people develop eating disorders because you massively overestimate their calories? If the goal is to learn then accurate information is important, although I suppose it's harder to get a VC to fund.

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

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

#226

Post by sumeno »

wnorris510 wrote: post #196
jraines wrote: post #191

The benefit of AI as a feature in calorie counting apps is great. But you have to track stuff manually for a while to realize the benefit. Specifically: using AI for restaurant meals is a convenient starting point, from which you use your hard-won feel for weight & caloric density of things to tweak the first pass to something closer to the mark. As of now I would not at all trust a fully AI calorie counting app

100%. You should not trust photo based calorie tracking and should only use it when there is no other realistic choice (e.g. eating out). You especially should not trust an app that does not do any research on the accuracy of it's algorithm or claims to be "consistently accurate" as many of them do currently. The truth is these algorithms have the potential to be accurate when values are averaged out over the long te…

> You especially should not trust an app that does not do any research on the accuracy of it's algorithm or claims to be "consistently accurate" as many of them do currently.

Your website says:

> SnapCalorie is the first app where you can take a picture of any meal and get an accurate calorie count and nutrition in seconds.

So which is it? Can you get an accurate calorie count of a meal from a picture or not?

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

#227

Post by vjulian »

kelstar18 wrote: post #132
vjulian wrote: post #105

I use ChatGPT (4o model) to track my daily macros, specifically calories, protein, fat, and carbs. I have a prompt containing my macro targets and a list of the foods I typically eat or have at home. Throughout the day, I report my meals, for example five large eggs, two slices of sprouted wheat toast, and a cup of spinach with a pat of butter. It calculates the macros for each meal and keeps a running macros total s…

Is it ironic that my AI detector is ringing alarm bells at this comment (both its overall cadence, as well as it’s particularly unnecessary use of an em dash)?

Not AI-written. Personally, I love em dashes and semicolons.

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

#228

Post by vjulian »

pacerwpg wrote: post #146
vjulian wrote: post #105

I use ChatGPT (4o model) to track my daily macros, specifically calories, protein, fat, and carbs. I have a prompt containing my macro targets and a list of the foods I typically eat or have at home. Throughout the day, I report my meals, for example five large eggs, two slices of sprouted wheat toast, and a cup of spinach with a pat of butter. It calculates the macros for each meal and keeps a running macros total s…

Cool, it can replace myfitnesspal features from 10 years ago, but do them worse. Revolutionary.

No, I find that app cumbersome.

To respond to others’ comments, I tend to eat whole foods. I don’t eat TV dinner or shop at Costco or such places.

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

#229

Post by vjulian »

Iulioh wrote: post #151
MangoToupe wrote: post #147

Earlier quoted context omitted.

Is the em-dash ever "necessary"? It was beloved by nerds on the internet long before there was any whisper of LLMs on the internet. What do you think chatbots are imitating if not humans who also use it?

I never saw a human use it in something different from literature. That's where I think the chatbots got it.

I write for a living and read literature at university. I find it odd that you’ve never encountered an em dash in common writing.

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

#230

Post by vjulian »

pacerwpg wrote: post #146
vjulian wrote: post #105

I use ChatGPT (4o model) to track my daily macros, specifically calories, protein, fat, and carbs. I have a prompt containing my macro targets and a list of the foods I typically eat or have at home. Throughout the day, I report my meals, for example five large eggs, two slices of sprouted wheat toast, and a cup of spinach with a pat of butter. It calculates the macros for each meal and keeps a running macros total s…

Cool, it can replace myfitnesspal features from 10 years ago, but do them worse. Revolutionary.

Also, do you mind dropping the sarcasm and explaining why you believe MyFitnessPal might be superior to ChatGPT in the usage I’m suggesting?
Post reply on HN