As the new year inspires millions to pursue healthier lifestyles, many struggle to achieve their weight loss goals. Nearly three-quarters of Americans are overweight, and obesity is on the rise globally, driving demand for costly interventions such as bariatric surgery and GLP-1 medications.
New research from MIT Sloan and Singapore Management University suggests a low-cost alternative: generative artificial intelligence (AI). The study found that AI tools providing personalized, real-time dietary guidance can help reduce body mass index (BMI), though they cannot replace the benefits of supportive communities like Weight Watchers.
The study, conducted by MIT Sloan marketing professor Catherine Tucker and SMU researcher Linyi Li, involved 416 participants over three weeks in late 2024. Participants used a generative AI tool that analyzed their meals and provided dietary recommendations. The results showed that the number of participants no longer classified as obese increased from six to 17—approximately 4% of the total.
“Weight loss is such a big challenge. If it were easy, everyone would do it,” Tucker said. “The fact that a digital tool like AI can have any effect is remarkable, especially compared to expensive interventions. It demonstrates cost-effective behavior change.”
Study Design: AI and Community Interaction
The researchers partnered with a Fortune 500 company in Asia that runs an online weight loss boot camp incorporating diet and exercise guidance via WeChat. Participants were divided into three groups:
- Group 1 (Control): Received general healthy-diet tips and access to the group chat but no AI tool.
- Group 2 (Private AI Analysis): Sent meal photos privately to receive AI-generated nutritional feedback.
- Group 3 (Public AI Analysis): Shared meal photos and AI analysis publicly within the group chat.
Key Findings
1. Generative AI Supports Weight Loss
Participants using the AI tool achieved greater weight loss than the control group. Average weight loss over three weeks was 0.966 kg in Group 1, 1.426 kg in Group 2, and 1.358 kg in Group 3. Tucker emphasized that AI excels in personalizing experiences and guiding users toward healthier choices.
2. Public AI Feedback Can Reduce Participation
Group 2, which received private AI feedback, had higher engagement and completion rates than Group 3, where feedback was public. Public sharing appeared to discourage some participants, leading to higher dropout rates. “Dropout is the big enemy of weight loss,” Tucker explained, noting that comparisons in a public setting can create pressure and reduce motivation.
3. AI May Reduce Health Inequalities
The research found that participants with less nutritional knowledge benefited most from personalized AI guidance. These individuals often struggle with conventional diet education, highlighting AI’s potential to improve health outcomes for underserved populations.
Implications
While generative AI can aid in individual weight loss through personalized recommendations, it cannot replace the sense of community that supports long-term behavior change. Tucker notes that ecosystems like Weight Watchers provide accountability, shared vulnerability, and social support that AI alone cannot replicate.
“Technology can assist with information retrieval, reminders, and prompts, but it cannot replace human connection,” Tucker said. The research suggests that firms should focus on AI’s personalization capabilities rather than attempting to build community solely through digital tools.
The study, “Building an Ecosystem or Prioritizing Personalization With AI? Evidence From a Field Experiment,” demonstrates that generative AI can be a valuable supplement to weight loss programs, offering low-cost, effective guidance while emphasizing the enduring importance of social support networks.
