What tools can do:
- Deepware Scanner: genuinely scans of suspicious videos, right in the browser.
- Reality Defender: 50 scans a month, no credit card, production-grade models.
- Hive: browser checker for quick image and video tests.
What they cannot do:
- Batch screening. Basic tiers are one file at a time. - Legal-grade reports. verdicts are indicative, not evidentiary. - Survive heavy compression reliably. A video recompressed three times by messaging apps degrades every detector. - API access. Automation starts at paid tiers. Sources: CISA online deepfake detection answers "is this specific video suspicious?" It does not answer "are any of our 10,000 daily uploads synthetic?" Know which question you are asking before you pick the tier.
Deepfake checkers are fine for curiosity; production decisions need measured accuracy, not a yes-or-no widget.