3,755 AI use cases where a classifier beats a chatbot

Every task here is a repeatable decision (yes/no, pick-from-a-list, or a 1–5 score) that a fast typed-decision model can answer cheaply and verifiably. Each one is graded S to X on fit, volume, value, and risk. Inspired by Jev from TypeSafe.

Browse the leaderboard Submit a use case JSON API

Use cases3,755

S-tier162

Seen in the wild1,903

Verticals19

Leaderboard

Top 100 S/A use cases

Vertical Function Status

Rank	Grade	Use case	Vertical	Function	Volume	Value	Risk	Score
1	S	Prompt injection detection guardrail	Software & tech	Agents & dev	high	major	high	5.94
2	S	Social media spam and abuse classification	Software & tech	Social & creator	high	major	moderate	5.92
3	S	Email spam/phishing detection	Cross-industry	Knowledge work	high	major	high	5.91
4	S	Jev System One AI model for instant decisions	Finance & insurance	Agents & dev	high	major	high	5.91
5	S	Content moderation unsafe tagging	Software & tech	Social & creator	high	major	high	5.90
6	S	Brand safety keyword classification	Agencies, media & creators	Ads	high	major	high	5.87
7	S	Chatbot guardrail layer for LLM concierges	Healthcare & dental	Agents & dev	high	major	high	5.87
8	S	Employee message pre-send insider-info guard	Finance & insurance	Legal & finance	high	major	high	5.87
9	S	Security alert categorization system	Cross-industry	Agents & dev	high	major	high	5.85
10	S	Privilege-risk flag on every outbound email	Legal	Legal & finance	high	major	high	5.82
11	S	Comment toxicity and harassment detection	Cross-industry	Social & creator	high	major	high	5.79
12	S	Content moderation and brand safety	Cross-industry	Social & creator	high	major	high	5.79
13	S	LLM output safety classification	Software & tech	Agents & dev	high	major	high	5.78
14	S	Minecraft chat moderation plugin blocks harassment and grooming	Games & entertainment	Social & creator	high	major	high	5.76
15	S	LLM cost-quality routing with RouteLLM	Software & tech	Agents & dev	high	major	moderate	5.75
16	S	Detect PHI in conversion events and URL parameters	Unclassified	Ads	high	major	high	5.72
17	S	Nyckel content moderation classifier	E-commerce & retail	E-commerce ops	high	major	high	5.69
18	S	Claude Code hooks pre-tool-use validation	Software & tech	Agents & dev	high	major	moderate	5.67
19	S	Log sensitive-data leak	Software & tech	Agents & dev	high	major	high	5.67
20	S	Search intent classification matching ad message	Cross-industry	Ads	high	major	moderate	5.67
21	S	E-commerce search query intent classification	E-commerce & retail	E-commerce ops	high	major	moderate	5.66
22	S	Cross-domain: sensitive-data leak guard on LLM prompts	Cross-industry	Agents & dev	high	major	high	5.64
23	S	Prompt injection scanner	Software & tech	Agents & dev	high	major	high	5.64
24	S	Claude Code agent hooks for guardrails	Software & tech	Agents & dev	high	major	high	5.63
25	S	Claude Code post-tool-use guardrail	Software & tech	Agents & dev	high	major	high	5.60
26	S	Detect toxic voice-chat transcripts for review	Games & entertainment	Social & creator	high	major	high	5.60
27	S	Pre-flag PHI and quasi-identifiers in content	Healthcare & dental	Agents & dev	high	major	high	5.59
28	S	Patient symptom urgency triage	Healthcare & dental	Health admin	high	major	high	5.58
29	S	Agent tool-call risk gate	Software & tech	Agents & dev	high	major	high	5.57
30	S	Escalation priority for trust-and-safety reports	Software & tech	Social & creator	high	major	high	5.55
31	S	Impersonation and fraud detection	Cross-industry	Legal & finance	high	major	high	5.55
32	S	Label client-level data for HIPAA-safe tool routing	Healthcare & dental	Agents & dev	high	major	high	5.55
33	S	Sentiment-based alert system	Cross-industry	Support & CX	high	major	moderate	5.54
34	S	Chat widget deflect vs hand to human	Healthcare & dental	Health admin	high	major	high	5.53
35	S	PMax placement brand safety	Cross-industry	Ads	high	major	high	5.53
36	S	Phishing triage	Cross-industry	Agents & dev	high	major	high	5.52
37	S	Detect restricted or regulated product in seller upload	E-commerce & retail	E-commerce ops	high	major	high	5.50
38	S	Free trial conversion prediction	Software & tech	Sales	high	major	moderate	5.49
39	S	E-discovery responsiveness first pass	Legal	Legal & finance	high	major	high	5.48
40	S	Churn risk prediction	Cross-industry	Support & CX	high	major	moderate	5.47
41	S	Policyholder chat: coverage promise guard	Finance & insurance	Legal & finance	high	major	high	5.47
42	S	Keystroke-level sensitive-text guard	Cross-industry	Agents & dev	high	major	high	5.44
43	S	Negative news detection reputation risk	Cross-industry	Local & reputation	high	major	high	5.43
44	S	Client sentiment risk	Cross-industry	Support & CX	high	major	moderate	5.42
45	S	Confidential message detector	Cross-industry	Agents & dev	high	major	high	5.41
46	S	Detect policy-violating posts in internal chat	Cross-industry	Ops & HR	high	major	moderate	5.39
47	S	Classify phishing vs legitimate emails	E-commerce & retail	Support & CX	routine	major	high	5.34
48	S	Detect PII or secrets pasted into support tickets	Cross-industry	Support & CX	routine	major	high	5.31
49	S	Insurance claim triage complexity	Finance & insurance	Legal & finance	high	major	high	5.31
50	S	Ad relevance quality score prediction	Cross-industry	Ads	high	major	moderate	5.30
51	S	Compliance linter for marketing copy	Cross-industry	Legal & finance	routine	major	high	5.30
52	S	Detect off-platform payment solicitation in messages	E-commerce & retail	E-commerce ops	routine	major	high	5.29
53	S	HIPAA-safe form sorting	Healthcare & dental	Health admin	routine	major	high	5.27
54	S	Patient diagnosis category	Healthcare & dental	Health admin	routine	major	high	5.27
55	S	Triage emergency vs scheduled job	Home & local services	Ops & HR	routine	major	high	5.26
56	S	PII in free text	Cross-industry	Agents & dev	high	major	high	5.25
57	S	Detect fake-review rings through review similarity	E-commerce & retail	Local & reputation	routine	major	moderate	5.24
58	S	Query intent for governance	Cross-industry	Agents & dev	routine	major	high	5.23
59	S	Detect counterfeit-risk language in listings	E-commerce & retail	E-commerce ops	routine	major	high	5.22
60	S	Detect fair-housing risky phrases in listing drafts	Real estate & construction	Legal & finance	routine	major	high	5.22
61	S	Detect when a seller is circumventing fees through coded language	E-commerce & retail	E-commerce ops	routine	major	moderate	5.22
62	S	Flag platform-policy risk in aesthetic social and ad creative	Software & tech	Ads	routine	major	high	5.22
63	S	Grid operator log scan for anomalies	Industrial, energy & agriculture	Creative & industry	routine	major	high	5.22
64	S	Battery storage fault description clustering	Industrial, energy & agriculture	Agents & dev	routine	major	high	5.21
65	S	Detect post-treatment concern in text message	Healthcare & dental	Health admin	routine	major	high	5.21
66	S	Detect patient-identifying detail in interview notes	Healthcare & dental	Health admin	routine	major	high	5.20
67	S	Complication language detector in inquiries	Healthcare & dental	Health admin	routine	major	high	5.19
68	S	Decide whether a page is YMYL and which blocks fire	Agencies, media & creators	SEO & AI search	routine	major	high	5.18
69	S	Cheating chat-log alert in remote proctoring	Education	Research & education	routine	major	moderate	5.16
70	S	Sort comments and DMs for lead monitoring	Healthcare & dental	Health admin	routine	major	high	5.16
71	S	Churn-risk language in support ticket threads	Software & tech	Support & CX	routine	major	moderate	5.14
72	S	HIPAA-safe form sorter	Healthcare & dental	Health admin	routine	major	high	5.13
73	S	Score ad-to-page message match	Cross-industry	Ads	routine	major	moderate	5.13
74	S	Invoice payment behavior credit risk	Finance & insurance	Legal & finance	routine	major	high	5.07
75	S	Outage ticket customer call text triage	Industrial, energy & agriculture	Support & CX	routine	major	high	5.07
76	S	Detect champion-leaving signals in call transcripts	Software & tech	Sales	routine	major	moderate	5.06
77	S	Day-of attendee message triage via chat app	Hospitality, travel & events	Creative & industry	routine	major	high	5.05
78	S	Social clip caption brand risk pre-check	Cross-industry	Social & creator	routine	major	high	5.05
79	S	Command-line suspiciousness	Cross-industry	Agents & dev	routine	major	moderate	5.03
80	S	PII scan before public access	Cross-industry	Agents & dev	routine	major	high	5.03
81	S	Rumor detection during disaster social posts	Government & nonprofit	Social & creator	routine	major	high	5.03
82	S	Red flag messaging and toxic copy detection	Cross-industry	Ads	routine	major	high	5.00
83	S	Ad-to-landing page message match	Cross-industry	Ads	routine	major	moderate	4.99
84	S	Medical appointment urgency	Healthcare & dental	Health admin	routine	major	high	4.98
85	S	Regulated-claim topic router	Cross-industry	Legal & finance	routine	major	high	4.98
86	S	Check legal copy against advertising and confidentiality rules	Legal	Legal & finance	routine	major	high	4.97
87	S	Pre-send guard on outbound email claims	Cross-industry	Email & CRM	routine	major	high	4.96
88	S	Before/after claim detection in copy	Cross-industry	Ads	routine	major	high	4.95
89	S	Is this agent draft reply safe to send	Cross-industry	Support & CX	routine	major	high	4.95
90	S	Negative press alert triage	Cross-industry	Local & reputation	routine	major	high	4.93
91	S	Personalized payment plan offer	Finance & insurance	Legal & finance	routine	major	high	4.93
92	S	Sort 311 requests free text to department	Government & nonprofit	Ops & HR	routine	major	high	4.90
93	S	Detect policy-risk ad copy for healthcare	Healthcare & dental	Ads	routine	major	high	4.89
94	S	FDA status claim detection	Healthcare & dental	Legal & finance	routine	major	high	4.89
95	S	Healthcare medical ad compliance checking	Healthcare & dental	Ads	routine	major	high	4.89
96	S	Customer satisfaction measurement	Cross-industry	Support & CX	routine	major	moderate	4.87
97	S	KYC document type confirmation	Finance & insurance	Legal & finance	routine	major	high	4.87
98	S	Auth log story	Cross-industry	Agents & dev	routine	major	moderate	4.86
99	S	Google healthcare policy risk in ad	Healthcare & dental	Ads	routine	major	high	4.83
100	S	Pitch deck email triage at production company	Games & entertainment	Creative & industry	routine	major	high	4.82

Explore by function

All functions ↗

↗Personalization & CRO150 use casesProduct recommendation ranking ↗Ads181 use casesBrand safety keyword classification ↗SEO & AI search277 use casesDecide whether a page is YMYL and which blocks fire ↗Local & reputation82 use casesNegative news detection reputation risk ↗Social & creator208 use casesSocial media spam and abuse classification ↗Email & CRM65 use casesPre-send guard on outbound email claims ↗Content & brand183 use casesDetect quantified efficacy claims needing substantiation ↗Sales318 use casesFree trial conversion prediction ↗Support & CX295 use casesSentiment-based alert system ↗E-commerce ops93 use casesNyckel content moderation classifier ↗Health admin74 use casesPatient symptom urgency triage ↗Legal & finance296 use casesEmployee message pre-send insider-info guard ↗Ops & HR155 use casesDetect policy-violating posts in internal chat ↗Agents & dev858 use casesPrompt injection detection guardrail ↗Research & education164 use casesCheating chat-log alert in remote proctoring ↗Creative & industry115 use casesGrid operator log scan for anomalies ↗Knowledge work194 use casesEmail spam/phishing detection ↗Other47 use casesConstituent email routing for a legislator

Grades by vertical

Vertical	S	A	B	C	X	?
Cross-industry	59	1027	50	457	127	5
Medical & aesthetic	3	28	1	1	0	0
Healthcare & dental	25	69	18	18	3	0
Legal	5	20	10	13	0	0
Home & local services	1	8	1	4	0	0
Real estate & construction	3	24	1	8	0	0
Agencies, media & creators	5	111	5	77	40	0
E-commerce & retail	10	72	1	23	15	0
Software & tech	23	249	22	271	198	5
Finance & insurance	10	48	7	33	16	0
Education	3	33	0	36	9	0
Hospitality, travel & events	2	23	0	11	0	0
Staffing & recruiting	0	51	0	16	0	0
Industrial, energy & agriculture	4	15	5	7	3	0
Logistics & supply chain	1	9	1	3	0	0
Government & nonprofit	2	29	3	20	0	0
Games & entertainment	4	27	2	51	26	0
Research & science	1	22	3	22	4	1
Unclassified	1	2	1	25	47	0

Cite or embed this directory

CC-friendly. Attribution link appreciated.

Reference the atlas in your article, deck, or agent prompt. Every use case has a stable URL, a markdown twin, and a JSON API.

Linkhttps://jevaiusecases.com/

HTML<a href="https://jevaiusecases.com/">Jev AI Use Cases: 3,755 graded classifier tasks</a>

Markdown[Jev AI Use Cases](https://jevaiusecases.com/)

JSON APIhttps://jevaiusecases.com/api/use-cases.json

Know a missing use case?

Submit a decision task for review, or use the agent guide.

Submit a use case ↗