Testing of detection tools for AI-generated text

Table 11 Logarithmic approach to accuracy evaluation

Tool	01-Hum	02-MT	03-AI	04-AI	05-ManEd	06-Para	Total	Accuracy	Rank
Check For AI	144	62	144	129	74	54	607	70%	7
Compilatio	136	144	136	132	91	40	679	79%	2
Content at Scale	144	144	23	24	17	18	370	43%	14
Crossplag	144	99	144	115	76	40	618	72%	6
DetectGPT	144	108	88	129	38	36	543	63%	10
Go Winston	124	124	144	130	79	45	646	75%	4
GPT Zero	102	60	121	128	89	89	589	68%	8
GPT-2 Output Detector Demo	144	114	144	129	84	35	650	75%	3
OpenAI Text Classifier	144	136	67	124	67	48	586	68%	9
PlagiarismCheck	128	108	76	82	50	53	497	58%	12
Turnitin	144	144	136	144	81	53	702	81%	1
Writeful GPT Detector	144	122	81	76	50	20	493	57%	13
Writer	144	117	83	84	53	35	516	60%	11
Zero GPT	144	108	120	132	65	54	623	72%	5
Average	96%	79%	75%	77%	45%	31%

ISSN: 1833-2595