Detect Python code written by AI easily with science

emoji_lulz.py

#!/usr/bin/env python3
import ast, re, sys

EMOJI_RE = re.compile(
    r"[\U0001F300-\U0001F5FF\U0001F600-\U0001F64F\U0001F680-\U0001F6FF"
    r"\U0001F700-\U0001F77F\U0001F780-\U0001F7FF\U0001F800-\U0001F8FF"
    r"\U0001F900-\U0001F9FF\U0001FA00-\U0001FA6F\U0001FA70-\U0001FAFF"
    r"\u2702-\u27B0\u24C2-\U0001F251\U00010000-\U0010FFFF]"
)

def lit_strings_from_print(call: ast.Call):
    if not (isinstance(call.func, ast.Name) and call.func.id == "print"):
        return ()
    for arg in call.args:
        if isinstance(arg, ast.Constant) and isinstance(arg.value, str):
            yield arg.value
        elif isinstance(arg, ast.Str):  # Py<3.8
            yield arg.s
        elif isinstance(arg, ast.JoinedStr):  # f-strings: only literal chunks
            for v in arg.values:
                if isinstance(v, ast.Constant) and isinstance(v.value, str):
                    yield v.value
                elif isinstance(v, ast.Str):
                    yield v.s

def analyze(path):
    code = open(path, "r", encoding="utf-8").read()
    tree = ast.parse(code)
    findings = []
    for node in ast.walk(tree):
        if isinstance(node, ast.Call):
            for s in lit_strings_from_print(node):
                n = len(EMOJI_RE.findall(s))
                if n:
                    findings.append({"line": getattr(node, "lineno", "?"), "string": s, "emoji_count": n})
    total = sum(f["emoji_count"] for f in findings)
    return total, findings

def main():
    if len(sys.argv) != 2:
        print("Usage: python emoji_detector.py <python_file.py>"); sys.exit(1)

    path = sys.argv[1]
    total, items = analyze(path)

    print(f"\nAnalyzing: {path}")
    print(f"Total emojis in print statements: {total}")
    if items:
        print("\nPrint statements with emojis:")
        for f in items:
            preview = (f["string"][:50] + "...") if len(f["string"]) > 50 else f["string"]
            print(f'  Line {f["line"]}: {f["emoji_count"]} emoji(s) - "{preview}"')
    print("\n⚠️  VERDICT: This code is likely AI-generated (>2 emojis in print statements)"
          if total > 2 else "\n✓ VERDICT: This code appears to be human-written (≤2 emojis in print statements)")

if __name__ == "__main__":
    main()