Terms of Use

Walmart · agreement dated September 9, 2026 · Official agreement

Read the agreement

2 of 92 cards in Security & Safeguards

Security & Safeguards · Restriction or exception

Prohibited conduct: harmful code, commercial use, harvesting, interference

You agree not to make available prohibited Content or harmful or invasive code, use the Sites for commercial or fraudulent, tortious or unlawful purposes, harvest or collect information about users, interfere with or disrupt the Sites or their systems, restrict others' use, cause injury, or violate any law or the Terms.

Supporting quotations

Make available any Content through or in connection with the Walmart Sites that is or may be in violation of the content guidelines set forth in Section 3.C (Prohibited Content) below.
virus, worm, Trojan horse, Easter egg, time bomb, spyware, or other computer code, file or program that is or is potentially harmful or invasive or intended to damage or hijack the operation of, or to monitor the use of, any hardware, software, or equipment
Use the Walmart Sites for any commercial purpose, or for any purpose that is fraudulent or otherwise tortious or unlawful.
Harvest or collect information about users of the Walmart Sites.
Interfere with or disrupt the operation of the Walmart Sites or the systems, servers, or networks used to make the Walmart Sites available
Restrict or inhibit any other person from using the Walmart Sites.
Cause injury to any person or entity.
Violate any law, rule, or regulation, or these Terms of Use.

purpose: security

Cross-references

Security & Safeguards · Restriction or exception

Generative AI features may not be used to cause harm or to circumvent safeguards

Using the Generative AI Features to produce violent, abusive or deceptive content or otherwise cause harm is strictly prohibited. You may also not attempt to “break” or “trick” the features into disregarding safeguards or other features or messages aimed to protect the public.

Supporting quotations

Using the Generative AI Features to produce violent, abusive, or deceptive content or to otherwise cause harm is strictly prohibited.
You are also prohibited from attempting to “break” or “trick” the Generative AI Features into disregarding safeguards or other features or messages aimed to protect the public.