In this Beta release, MacroFactor AI allows you to capture a meal with a photo and automatically populate your Plate with nutrient complete, editable food entries. It’s designed to help both experienced MFers save time and make food logging more accessible to new users.
Keep in mind that the Beta feature is a work in progress. The next generation of MacroFactor AI is already underway and outperforming the Beta in every capacity.
MacroFactor is focused on intentional AI
With the excitement surrounding AI, hundreds of apps have flooded the market with LLM-based food loggers. These apps almost universally lack the depth and functionality necessary to be complete solutions for the dedicated user.
MacroFactor offers a more serious alternative with intentional application of AI. Our focus is on:
- High quality results. MacroFactor AI does not rely on LLMs to generate all food entries. Instead, we prioritize real, lab-analyzed results with complete nutrition data. MacroFactor AI will only generate a food entry when it is necessary or otherwise optimal, and its generation will be grounded in real life foods.
- Result transparency. MacroFactor AI will break down your meal into individual ingredients instead of giving you a single opaque result. Because we build each result from individual food entries, our output is fully inspectable.
- Integrated control: The plate is a unified interface for interacting with the AI result and human-added foods. AI is not perfect and can make mistakes, and we give users the control to edit and modify its results.
- Streamlined experience: MacroFactor AI will do its best to group foods logically, using recipes and single foods, so it is easy for you to reason about your nutrition. Our AI interface has no fluff, no performative chat interface – just streamlined results.
Our intention is that AI makes you more powerful and consistent in your logging habits. In turn, MacroFactor’s advanced analytics and diet coaching work better when you are more consistent. MacroFactor analyzes your metabolism, guides you through your goals like a macro coach would, and gives you a complete picture of your nutrition – calories, macros, and micros. Photo-based logging is one of seven food-logging workflows in MacroFactor that was designed carefully, so you can log foods faster and take better advantage of our analytics and coaching guidance.
Our near-term goal is that MacroFactor AI will:
- Save veteran users time by automating the bulk search process for complex dishes.
- Perform better than an untrained human, removing barriers to entry for those who are new to food logging.
- Bridge the experiential gap for those who live in regions where food and/or branded product data is not as readily available.
Our long-term goal is that MacroFactor AI will be an ecosystem of utilities that will nearly eliminate food logging effort for even the most experienced MFer, without compromising on results, transparency, or end-user control.
How to use photo-based AI to log foods in MacroFactor
Navigate to “AI” from MacroFactor’s food logger. For the fastest workflow, simply snap a photo. If you’d like to add more context, choose the “Photo & Text” option and add a quick description of your meal alongside the photo.
From here, AI will go to work. Behind the scenes, several LLM prompts will work in tandem to break down your meal into individual ingredients and recipes, querying our database to retrieve real foods.
Once your meal is fully analyzed, foods will stream into your plate. Results might be served as individual foods or be intelligently grouped into recipes as shown in the example below.
From here, you can inspect any result produced by the MacroFactor AI. Your plate has an in-context editor for each food, including recipe ingredients, so you can tweak AI’s result and make it perfect prior to logging.
If AI misses something, tap on one of the utilities on the bottom of your plate. You can continue scanning, searching, and even using AI again to add more foods. The advantage of MacroFactor’s unified food logging interface is that you can work collaboratively with AI, taking advantage of AI for speed without sacrificing control.
Using AI without sacrificing accuracy
If you are new to food logging, AI might perform with greater accuracy than you at logging certain meals, especially once MacroFactor AI is out of Beta. For example, when logging restaurant meals or foods prepared outside the home, it’s perfectly normal for beginners to forget to account for sauces and have a harder time estimating portions. Even when AI makes mistakes, it will enable you to log consistently and ramp up quickly. Food logging is a skill and takes logging reps to develop.
However, we know that many veteran food loggers value accuracy above all else — and some may be skeptical that an AI will meet their standards. For example, your burger might have a perfectly hidden layer of sauce under its bun. Sometimes, AI may not have enough context or might make unintended mistakes.
We have full confidence that by the end of the year, MacroFactor AI will reach a level of performance that even veterans will readily rely on. But even now, MacroFactor AI isn’t about trading precision for convenience. It’s designed to save you time without taking away your control. AI can reduce the effort of searching and selecting ingredients — especially in complex meals with lots of components.
For example, a dish might contain as many as 11 ingredients and be made of 3 or more constituent recipes. Previously, logging this meal would require 11 unique scans or searches and advanced use of recipe features. Now, AI might eliminate most of that work. Moreover, AI will intelligently group results for you, so you do not need to create recipes to have the benefit of a tidy food logging result.
You can still review, adjust, and confirm every entry before logging, taking advantage of AI as a speedy collaborator rather than the primary decision maker. In fact, we recommend that everyone reviews their AI results before logging.
Remember, MacroFactor does not just adapt to your metabolism, but also to your food logging habits. Its algorithms can gracefully subsume food logging errors and still guide you to your goals.
4 real examples of MacroFactor AI at work
1. Logging a plate of assorted branded products.
When scanning a single branded product, using a barcode scanner is often faster. But if you are eating a few foods at once, AI can speed up your workflow by doing the searches for you. MacroFactor AI will identify the branded products in your meal and search our product database. In this case, instead of prioritizing micronutrient-rich common foods, AI will look for products that match the brand on display.
Take a look at the example above. Usually, logging this yogurt, drink, and protein bar would require three consecutive barcode scans or manual searches. MacroFactor AI can save you time and perform these operations for you.
Using the photo of the three products, MacroFactor AI correctly identified a Coca-Cola drink and that it is the zero sugar variant, and it saw that Fage is a full-fat yogurt at 5%. It also found the Power Crunch bar.
2. Logging a waffle after an Uber Eats delivery.
MacroFactor AI can also identify multiple foods from your deliveries, even if they are not explicitly branded. In this case, MacroFactor AI will prioritize building your result from research-grade, micronutrient-rich food entries.
Taking a look at the waffle dish below, AI added a Waffle, a Fruit Salad, and Chocolate Syrup to the plate.
The identification of foods is in line with expectation, but when logging, you might prefer to transfer the fruits to the waffle and treat them as individual toppings rather than an aggregated Fruit Salad.
You can split the recipe into individual foods by swiping on the Fruit Salad card and tapping the Explode icon. This action will extract Strawberries, Bananas, and Blueberries from the recipe and put them directly into your plate.
Now you can log the meal to commit it to your timeline!
AI helps you identify the foods and reason about servings, but you have control to make the result your own.
3. Logging a homemade brunch.
When you cook at home, you have the opportunity to log your meals with excellent detail. You know the ingredients you used, the spices you sprinkled, and the cooking fat you added. However, logging your home-cooked meals can also present the greatest challenge due to the sheer number of ingredients involved. For example, even a simple brunch plate below contains 10 ingredients: bread, eggs, cheese, three kinds of sweet potato, broccoli, olive oil, salt, and pepper.
MacroFactor AI broke down this dish into three distinct recipes: fried egg sandwich, steamed broccoli, and roasted root vegetables. Each of these recipes contains the ingredients you see in the image. Let’s inspect the star of the dish, the fried egg sandwich, further.
At first, there is a red flag: AI only detected one sandwich. Fortunately, upon further inspection, you can see that AI actually recognized the correct ingredients, simply bundling them under one recipe. 2x white bread, 1x cheese slice, 2x egg, salt, and pepper were all detected with correct serving sizes.
One advantage of using MacroFactor AI is that it can often take into account little details that a human would usually not bother to log, such as salt. While salt does not move your calories, it contributes to the micronutrient reporting you can access in MacroFactor’s Nutrient Explorer.
4. Logging a complex meal from a hibachi restaurant.
Eating out often presents a significant challenge for food loggers. Unlike with homemade dishes, it is difficult to know the precise amount of cooking fat used, the serving sizes for ingredients, and the impact of the sauces on your caloric totals.
If you are eating out in a social setting, it can also be challenging to fully account for your meal without detracting from the experience. MacroFactor AI can leverage its knowledge of different cuisines and help you log your meal quickly.
The dish in the above photo is Hibachi Chicken from a restaurant in North Carolina. MacroFactor AI identified three major components of the plate: Chicken Teriyaki, White Rice, and Steamed Broccoli. When inspecting the Chicken Teriyaki, you can see that AI was detailed in its analysis, accounting for the protein, sauce, sesame oil, and seasonings often used in this dish.
A seasoned operator might notice that the serving for rice seems larger than identified. AI is not perfect and can make mistakes, but fortunately, in this case MacroFactor identified the foods and the serving units correctly, so you can adjust the quantity of the rice from 1 cup to 2 cups using an inline plate editor.
Now you can log your meal and get back to the conversion.
10 tips to become an AI food logging pro
1. Light it up. Good lighting means better results. Snap your meal in bright, even light to help the AI clearly see what’s on your plate.
2. Make sure ingredients are fully in view. Angle your camera such that ingredients on your plate are fully visible. Usually, this means taking a photo top down.
3. Keep it tidy. The clearer the plate layout, the better. Spread your foods out so they’re not stacked or hidden — it helps the AI recognize each item correctly.
4. You can add a quick description. A short meal description (“chicken curry with rice and naan”) can boost accuracy, especially for complex or regional dishes.
5. Double-check and adjust. AI can significantly speed up your search by identifying all visible ingredients, but you’re the expert on what’s actually on your plate. Quickly inspect the entries, tweak portions if needed, and log. Even if AI does not get you 100% of the way there, it will save you many taps.
6. Know your use case. If you’re having a protein bar, packaged snack, or branded product, it may be faster to log it with the barcode scanner. AI is most efficient with mixed meals, plates, and homemade dishes.
7. Include your fist for scale. You can include your fist or a common object for scale to help MacroFactor AI estimate portions.
8. You can use an actual scale. If you are weighing out your food, take a photo with the scale in view. This can help AI reference the total weight of your meal.
9. Watch out for updates. MacroFactor AI will continue evolving. We expect the next generation model to provide a significant boost to both ingredient and serving size estimations.
10. Join the MacroFactor Reddit and Facebook communities. See how others leverage AI and get advice from experienced users and the MacroFactor team.
What will you log?
Show us your meals in the MacroFactor app community on Facebook and the MacroFactor subreddit.
If you are ready to get started with MacroFactor, you can download the app now from the App Store or Google Play.
You can also learn more about macro tracking, nutrition, and physiology by reading our in-depth articles or browsing our extensive knowledge base.
