You may occasionally find that Zoya’s Shariah compliance result differs from ChatGPT, Gemini, another screening app, or another source online. There are a few common reasons why.
AI tools are not definitive Shariah screening sources
General-purpose AI tools can be useful for research, but they are not dedicated Shariah screening systems.
They may rely on outdated information, use a different screening methodology, or pull from sources that conflict with each other.
They can also miss important details buried in SEC filings, annual reports, footnotes, segment disclosures, or other financial documents that may affect a company’s compliance status.
For that reason, an AI-generated answer should not be treated as a definitive compliance ruling on its own.
Different Shariah standards can produce different results
There is no single universal screening methodology.
Standards such as AAOIFI, S&P Shariah, MSCI Islamic, Dow Jones Islamic, and FTSE differ in how they calculate financial ratios, apply thresholds, and assess certain business activities.
That means the same company can receive different results depending on the methodology being used.
Zoya shows which screening standard is applied to each report.
Compliance can change over time
A company can move from compliant to non-compliant, or vice versa, as its financials and business activities change.
Zoya updates its screening data as new financial information becomes available. Another source may still be using an older classification.
How does Zoya screen stocks?
Zoya applies established Shariah screening methodologies using company financials and business activity data. Our screening process is also overseen by our Shariah board who provide guidance on how Islamic principles should be applied, particularly in cases that require judgment beyond simple financial ratios. We also show the underlying screening report so you can see why a company passed or failed a particular criterion.
What if another source disagrees with Zoya?
Send us the company name or ticker, along with the source showing a different result. We’re happy to review it. Differences often come down to methodology, data timing, or information one source considered that another did not.
If there is an issue with our own data or classification, we want to know about that too.