Post-market complaint tracking and trending; Vigilance data; Commercialization history
The sponsor utilized post-market performance data, including complaint tracking and vigilance reports, to support the effectiveness of the remanufactured AED batteries.
Post-market surveillance; Complaint tracking; Vigilance data
Clinical Evidence
Study Design
Population
Comparator
Key Endpoints
Post-market surveillance and commercialization history; Retrospective analysis of post-market performance; Follow-up/Duration: Since 2012; Study Period: 2012-2021
Users of AED Battery Exchange models in the US market; Number of Sites: Not applicable
Not applicable for this study
Safety and effectiveness based on complaint tracking and vigilance
Indications for Use
The automated external defibrillator (AED) battery supplies power to an AED as required during self maintenance, automated diagnoses, and defibrillation. The 9146-ABE is indicated for use with the Cardiac Science Powerheart G3, models 9390A, 9390E, 9300A, and 9300E. The G5-ABE is indicated for use with the Cardiac Science Powerheart G5, models G5A-80A, G5A-80C, G5S-80A, and G5S-80C. The 5070-ABE is indicated for use with the Philips HeartStart OnSite/Home, models M5066A, M5068A, and the FRx, model 861304. The FR3-ABE is indicated for use with the Philips HeartStart FR3, models 861388 and 861389.
Device Story
Remanufactured (re-celled) primary non-rechargeable batteries for specific AED models. Input: N/A (power source). Operation: Provides electrical power to AEDs for self-maintenance, automated diagnostics, and defibrillation. Used in clinical and public settings by trained personnel or lay-persons to treat sudden cardiac arrest. Output: Electrical power. Benefit: Enables AED functionality for life-saving defibrillation therapy. Safety verified through bench testing and compliance with OEM specifications.
Clinical Evidence
No clinical data submitted. Safety and effectiveness supported by in-vitro bench testing, literature, and clinical data previously submitted for compatible AEDs (P160028, P160029, P160033). Bench testing verified compliance with IEC 60601-2-4, IEC 60601-1, and UN38.3 standards.
Indicated for use in compatible AEDs to provide power for self-maintenance, automated diagnostics, and defibrillation therapy for victims of sudden cardiac arrest. Compatible with specific Cardiac Science Powerheart and Philips HeartStart models. Contraindicated for patients who are conscious or breathing normally.
Regulatory Classification
Identification
An automated external defibrillator (AED) system consists of an AED and those accessories necessary for the AED to detect and interpret an electrocardiogram and deliver an electrical shock (e.g., battery, pad electrode, adapter, and hardware key for pediatric use). An AED system analyzes the patient's electrocardiogram, interprets the cardiac rhythm, and automatically delivers an electrical shock (fully automated AED), or advises the user to deliver the shock (semi-automated or shock advisory AED) to treat ventricular fibrillation or pulseless ventricular tachycardia.
{0}
PMA P190013: FDA Summary of Safety and Effectiveness Data
# SUMMARY OF SAFETY AND EFFECTIVENESS DATA (SSED)
## I. GENERAL INFORMATION
Device Generic Name: Automated External Defibrillator
Device Trade Name: AED Battery Exchange (Models 9146-ABE, G5-ABE, 5070-ABE, FR3-ABE)
Device Procode: MKJ, NSA
Applicant's Name and Address: AED Battery Exchange, LLC
1000 Brown Street, Suite 206
Wauconda, Illinois 60084
Date(s) of Panel Recommendation: None
Premarket Approval Application (PMA) Number: P190013
Date of FDA Notice of Approval: February 2, 2021
AED Battery Exchange, LLC remanufactured (or "re-celled") batteries have been on the market in the US since 2012, when FDA issued clearance through the 510(k) Premarket Notification process (K120350, "Extended Life Battery"). Compatible AEDs at the time of K120350 clearance included the Cardiac Science Powerheart AED G3 and the Philips FRx, Onsite/Home, and FR3. P190013 has been submitted in response to the Final Order issued January 29, 2015, in the Federal Register Volume 80 Number 19, Docket No. FDA-2013-N-0234 and republished February 3, 2015, in the Federal Register Volume 80 Number 22, Docket No. FDA-2013-N-0234. This Order required premarket approval of marketed pre-amendment Class III Automated External Defibrillator (AEDs) and Accessories (e.g., batteries, electrode pads). Products affected by this Order are the AED Battery Exchange 9146-ABE and 5070-ABE. Please note that the AED Battery Exchange G5-ABE and FR3-ABE were not pre-amendment Class III AED accessories and are requesting marketing approval for the first time in P190013. A combination of relevant literature and in-vitro bench testing has been reviewed to demonstrate a reasonable assurance of safety and effectiveness for the AED Battery Exchange, models 9146-ABE, G5-ABE, 5070-ABE, and FR3-ABE.
## II. INDICATIONS FOR USE
The automated external defibrillator (AED) battery supplies power to an AED as required during self maintenance, automated diagnoses, and defibrillation. The 9146-ABE is indicated for use with the Cardiac Science Powerheart G3, models 9390A, 9390E, 9300A, and 9300E. The G5-ABE is indicated for use with the Cardiac Science Powerheart G5, models G5A-80A, G5A-80C, G5S-80A, and G5S-80C. The 5070-ABE is indicated for use with the Philips HeartStart OnSite/Home, models M5066A, M5068A, and the FRx, model
{1}
861304. The FR3-ABE is indicated for use with the Philips HeartStart FR3, models 861388 and 861389.
## III. CONTRAINDICATIONS
Automated external defibrillators should not be used when a patient is conscious or breathing normally.
## IV. WARNINGS AND PRECAUTIONS
The warnings and precautions can be found in the AED Battery Exchange labeling.
## V. DEVICE DESCRIPTION
The 9146-ABE, G5-ABE, 5070-ABE, and FR3-ABE are primary batteries (non-rechargeable) that provide power to AEDs during self maintenance, automated diagnoses and defibrillation. Table 1 lists the four (4) battery models and their compatible AEDs.
Table 1: AED Battery Exchange (Models 9146-ABE, G5-ABE, 5070-ABE, and FR3-ABE)
| | Compatible AEDs | Description |
| --- | --- | --- |
| 9146-ABE | Cardiac Science™ Powerheart® G3 models 9390A, 9390E, 9300A, 9300E (P160033) | Lithium-Sulfur Dioxide (Li-SO2), 12 V, 7.5 A h |
| G5-ABE | G5-ABE: Cardiac Science™ Powerheart® G5 models G5A-80A, G5A-80C, G5S-80A, G5S-80C (P160033) | Lithium-Sulfur Dioxide (Li-SO2), 12 V, 7.5 A h |
| 5070-ABE | Philips™ HeartStart™ Onsite/Home models M5066A, M5068A (P160029); FRx model 861304 (P160028) | Lithium Manganese Dioxide (LMnO2), 9 V, 4.2 A h |
PMA P190013: FDA Summary of Safety and Effectiveness Data
Page 2
{2}
| | Compatible AEDs | Description |
| --- | --- | --- |
| FR3-ABE | Philips™ HeartStart™
FR3 models 861388 and
861389 (P180028) | Lithium Manganese
Dioxide (LMnO2), 9 V,
4.2 A h |
AED batteries must be installed in a compatible AED before they can be used. When installed, compatible AEDs are intended for use by personnel trained in their operation (Cardiac Science Powerheart® G3 models 9390A, 9390E, 9300A, 9300E; Cardiac Science™ Powerheart® G5 models G5A-80A, G5A-80C, G5S-80A, G5S-80C); users who have received training basic life support/AED or as part of a physician-authorized emergency medical response training program (Philips™ HeartStart™ FRx model 861304); users who have received training in Basic Life Support (BLS), Advanced Life Support (ALS), or another physician-authorized training program (Philips™ HeartStart™ FR3 models 861388 and 861389); and lay-persons (Philips™ HeartStart™ Onsite/Home models M5066A, M5068A) to treat victims of sudden cardiac arrest.
## VI. ALTERNATIVE PRACTICES AND PROCEDURES
Defibrillation is the only currently available treatment for termination of ventricular fibrillation (VF) or pulseless ventricular tachycardia (VT). Public access defibrillation is designed to provide potentially lifesaving treatment prior to the arrival of emergency personnel.
## VII. MARKETING HISTORY
The 9146-ABE and 5070-ABE have been commercially available in the US since 2012, cleared under K120350. The G5-ABE and FR3-ABE have not been marketed in the United States or any foreign country.
## VIII. POTENTIAL ADVERSE EFFECTS OF THE DEVICE ON HEALTH
Below is a list of the potential adverse effects (e.g., complications) associated with the use of the device and AEDs in general, listed in decreasing order of seriousness:
- Failure to identify shockable arrhythmia;
- Failure to deliver a defibrillation shock in the presence of VF or pulseless VT, which may result in death or permanent injury;
- Inappropriate energy which could cause failed defibrillation or post-shock dysfunction;
- Myocardial damage;
PMA P190013: FDA Summary of Safety and Effectiveness Data
{3}
- Fire hazard in the presence of high oxygen concentration or flammable anesthetic agents;
- Electromagnetic interference (EMI) from the defibrillator impacting other devices especially during charge and energy transfers;
- Incorrectly shocking a pulse sustaining rhythm and inducing VF or cardiac arrest;
- Bystander shock from patient contact during defibrillation shock;
- Interaction with pacemakers;
- Skin burns around the electrode placement area;
- Allergic dermatitis due to sensitivity to materials used in electrode construction; and
- Minor skin rash.
## IX. SUMMARY OF NONCLINICAL STUDIES
### Bench Testing
Through risk assessment conducted as part of a risk management process in alignment with the FDA-recognized voluntary consensus standard ISO 14971:2019, relevant hazards and estimated risks associated with the specific manufacturing procedure for remanufacturing the batteries were identified, in consideration of the performance of the original equipment manufacturer (OEM) batteries. Risk control measures were then identified and verified through non-clinical bench performance testing, which included selective application of relevant clauses from FDA-recognized voluntary consensus standards and other consensus standards. Table 2 summarizes design verification and design validation activities conducted to demonstrate a reasonable assurance of safety and effectiveness for the AED Battery Exchange (Models 9146-ABE, G5-ABE, 5070-ABE, FR3-ABE).
Table 1: Bench Testing
| Test Title | Acceptance Criteria | Results |
| --- | --- | --- |
| Medical electrical equipment - Part 2-4: Particular requirements for basic safety and essential performance of cardiac defibrillators (IEC 60601-2-4:2010/(R)2015) | Acceptance criteria as specified by Subclause 201.101.2 – Requirements for INFREQUENT USE, MANUAL DEFIBRILLATORS and Subclause 201.1.2.2 – Requirements for MANUAL DEFIBRILLATORS | Acceptable |
| Drop Test | Conducted in accordance with IEC 60601-1:2012 and in consideration of ANSI/AAMI ES60601-1:2005 + A1:2012, Subclause 15.3.4. Tests | Acceptable |
PMA P190013: FDA Summary of Safety and Effectiveness Data
{4}
| Test Title | Acceptance Criteria | Results |
| --- | --- | --- |
| | conducted within OEM AED and as a stand-alone battery. | |
| Shock Cycle Count Verification | Minimum shock cycle count verified using specifications of OEM batteries. | Pass |
| Operation Time Verification | Minimum operation time verified using specifications of OEM batteries. | Pass |
| Storage Temperature Verification | Storage temperature verified using specifications of OEM batteries. | Pass |
| Charge State Recognition | Charge state recognition verified for conformity with IEC 60601-2-4:2010/(R)2015, Subclause 201.15.4.3.101 using OEM AED. | Pass |
| Label Integrity Verification | Conducted in accordance with IEC 60601-1:2012 and in consideration of ANSI/AAMI ES60601-1:2005 + A1:2012 | Pass |
| Software Verification | Conducted in accordance with FDA guidance, “Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices.” | Acceptable |
| Enclosure Protection (IEC 60529) | Ingress Protection (IP) ratings evaluated using specifications of OEM batteries. | Pass |
| EMC Verification by Component Analysis | N/A | Acceptable |
| Shelf Life Verification by Component Analysis | N/A | Acceptable |
PMA P190013: FDA Summary of Safety and Effectiveness Data
{5}
| Test Title | Acceptance Criteria | Results |
| --- | --- | --- |
| Thermal Extreme (UN38.3) | Conducted in accordance with “Recommendations on the Transport of Dangerous Goods, Manual of Tests and Criteria,” Section 38.3. | Pass |
| Vibration (UN38.3) | Conducted in accordance with “Recommendations on the Transport of Dangerous Goods, Manual of Tests and Criteria,” Section 38.3. | Pass |
| Shock Safety (UN38.3) | Conducted in accordance with “Recommendations on the Transport of Dangerous Goods, Manual of Tests and Criteria,” Section 38.3. | Pass |
| External Short Circuit Safety (UN38.3) | Conducted in accordance with “Recommendations on the Transport of Dangerous Goods, Manual of Tests and Criteria,” Section 38.3. | Pass |
| Altitude Safety (UN38.3) | Conducted in accordance with “Recommendations on the Transport of Dangerous Goods, Manual of Tests and Criteria,” Section 38.3. | Pass |
| Label Integrity Verification | Conducted in accordance with IEC 60601-1:2012. | Pass |
| Software Verification | Conducted in accordance with FDA guidance, “Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices,” issued May 2005. | Pass |
PMA P190013: FDA Summary of Safety and Effectiveness Data
Page 6
{6}
| Test Title | Acceptance Criteria | Results |
| --- | --- | --- |
| Packaging (UN38.3 tested), 49 CFR compliant for US ground transportation | Conducted in accordance with “Recommendations on the Transport of Dangerous Goods, Manual of Tests and Criteria,” Section 38.3. | Pass |
## X. SUMMARY OF PRIMARY CLINICAL STUDIES
The final order, Effective Date of Requirement for Premarket Approval for Automated External Defibrillator Systems, published on January 29, 2015 and republished on February 3, 2015, states that clinical study information can be leveraged for AEDs from both published studies and clinical data previously submitted to FDA under the 510(k) Premarket Notification process. AED Battery Exchange, LLC did not submit clinical data in support of a reasonable assurance of safety and effectiveness. Data supportive of a reasonable assurance of safety and effectiveness is available in the compatible AED Summaries of Safety and Effectiveness Data:
P160028: https://www.accessdata.fda.gov/cdrh_docs/pdf16/P160028B.pdf
P160029: https://www.accessdata.fda.gov/cdrh_docs/pdf16/P160029B.pdf
P160033: https://www.accessdata.fda.gov/cdrh_docs/pdf16/P160033B.pdf
## XI. PANEL MEETING RECOMMENDATION AND FDA'S POST-PANEL ACTION
In accordance with the provisions of section 515(c)(3) of the act as amended by the Safe Medical Devices Act of 1990, this PMA was not referred to the Circulatory Systems Devices Panel, an FDA advisory committee, for review and recommendation because the information in the PMA substantially duplicates information previously reviewed by this panel on January 25, 2011 as part of the 515(i) process. The majority of the panel recommended that AEDs be regulated as Class III PMAs to have better oversight of device manufacturing and post-market performance.
## XII. CONCLUSIONS DRAWN FROM PRECLINICAL AND CLINICAL STUDIES
### A. Effectiveness Conclusions
Benefits of use of the batteries with their specific AEDs outweigh the risks with which it is associated. This is verified through the longevity of the commercialization of the products, through the literature on clinical safety and effectiveness, and through the post-market data, including tracking and trending of complaints and vigilance.
PMA P190013: FDA Summary of Safety and Effectiveness Data
{7}
B. Safety Conclusions
The nonclinical testing and data collected on the battery models 9146-ABE, 5070-ABE, G5-ABE, and FR3-ABE in use with their respective AEDs did not identify or result in any unacceptable safety concerns associated with use of the AED Battery Exchange Models 9146-ABE, 5070-ABE, G5-ABE, and FR3-ABE.
C. Benefit-Risk Determination
The benefit of early defibrillation therapy is the survival of patients in sudden cardiac arrest (SCA). The time from collapse to defibrillation is critical in patient survival. For every minute that passes between collapse and defibrillation, survival rates from ventricular fibrillation (VF) SCA decrease by 7% - 10%.
The magnitude of this benefit is either life or death. Patients are likely to put a high value on this treatment because it has the potential to save their lives. Patients are therefore willing to accept the risks of this treatment to achieve the benefit. If the treatment provides timely successful defibrillation, the patient will (or may) survive a life-threatening cardiac arrest situation and will be able to seek further treatment.
1. Patient Perspectives: This submission did not include specific information on patient perspectives for this device.
In conclusion, given the available information above, the data support that for the AED Battery Exchange 9146-ABE (when used with the Cardiac Science Powerheart G3, models 9390A, 9390E, 9300A, and 9300E); the AED Battery Exchange G5-ABE (when used with the Cardiac Science Powerheart G5, models G5A-80A, G5A-80C, G5S-80A, and G5S-80C); the AED Battery Exchange 5070-ABE (when used with the Philips HeartStart OnSite/Home, models M5066A, M5068A, and the FRx, model 861304); and the AED Battery Exchange FR3-ABE (when used with the Philips HeartStart FR3, models 861388 and 861389), the probable benefits outweigh the probable risks.
D. Overall Conclusions
The data in this application support the reasonable assurance of safety and effectiveness of the AED Battery Exchange 9146-ABE (when used with the Cardiac Science Powerheart G3, models 9390A, 9390E, 9300A, and 9300E); the AED Battery Exchange G5-ABE (when used with the Cardiac Science Powerheart G5, models G5A-80A, G5A-80C, G5S-80A, and G5S-80C); the AED Battery Exchange 5070-ABE (when used with the Philips HeartStart OnSite/Home, models M5066A, M5068A, and the FRx, model 861304); and the AED Battery Exchange FR3-ABE (when used with the Philips HeartStart FR3, models 861388 and 861389), respectively, when used in accordance with the indications for use.
PMA P190013: FDA Summary of Safety and Effectiveness Data
Page 8
{8}
PMA P190013: FDA Summary of Safety and Effectiveness Data
Page 9
XIII. **CDRH DECISION**
CDRH issued an approval order on February 2, 2021.
XIV. **APPROVAL SPECIFICATIONS**
Directions for use: See device labeling.
Hazards to Health from Use of the Device: See Indications, Contraindications, Warnings, Precautions, and Adverse Events in the device labeling.
Post-approval Requirements and Restrictions: See approval order.
Two short videos show you everything — or skip straight to the written tutorial if you'd rather read. You can reopen this any time from the Tutorial button in the top bar.
Part 1 — Search, results, and everyday workflows 16 min
Part 2 — Embeddings: the galaxy map 3 min
1. Search: exact and fuzzy
Type a phrase like "coronary artery calcification" into the search box. You get two kinds of results. Exact results match the literal phrase — prefix searches work ("coronary artery calcificati") but suffix searches do not. Fuzzy results match on the meaning and intent of your phrase rather than the exact words, and are sorted by relevance score. Hover over the Exact or Fuzzy badge on any row to see exactly why it matched.
Use the checkboxes above the results to narrow: SaMD keeps only software-only devices, AI / ML keeps only devices with AI.
Exact vs. fuzzy search: what's the difference?
Exact matches on the literal phrase (prefix search works, suffix does not). Fuzzy matches on the meaning and intent of the phrase rather than the exact words. Hover over the badge on any row to see why it matched.
You search "coronary artery calcification" and want only software devices with AI. What two filters do you apply?
Narrow by SaMD (software-only devices), then narrow by AI/ML (devices with AI).
2. The results table
Scroll right in the results table. The intended use is extracted for you — no need to open the PDF. The device story gives a high-level snapshot of what the device does and how it's used. The AI Performance sub-table shows each output name, acceptance criteria, observed values, and development/test dataset descriptions — the same format Innolitics uses for regulatory strategy outputs, and the fastest high-level fingerprint of an AI device. It is AI-generated but has been very reliable in practice.
Where do you find a device's intended use without opening the PDF?
Scroll right in the search results table. The intended use column is extracted for you; no need to dig into the 510(k) summary PDF.
What does the AI Performance sub-table show, and why is it useful?
Output name, acceptance criteria, observed values, development dataset description, and test dataset description. It's the same format we use for regulatory strategy output and Fast 510(k) input, and the fastest high-level fingerprint of an AI device. AI-generated but reliable in practice.
3. Judging fuzzy relevance
Fuzzy results trail off in relevance as you scroll. Use three signals to decide how far down to go: the fuzzy badge explanations, the intended use column, and whether your target output (e.g., Cobb angle) still appears in the AI Performance sub-table. Once it stops appearing, you're past the relevant zone. A top hit with a low score (~0.4) and a stretched explanation is a hint the closest predicates are far away — the project may be headed for De Novo. Note the fuzzy search is a pattern match: it doesn't handle negation ("not") well, and hardware devices can appear — filter by SaMD/AI ML to cut them.
How do you judge how far down fuzzy search results to go?
Use the relevancy signals: the fuzzy badge explanations, the intended use column, and whether the target output (e.g., Cobb angle) still appears in the AI Performance sub-table. Once it stops appearing, results are trailing off in relevancy.
4. Device detail page: chat and citations
Click a device name to open its detail page: device facts on the left, a chat window on the right. Ask something like "Describe the training data". The answer carries little citation bubbles — click one to jump to the highlighted passage in the source PDF, so you can verify every AI answer against the document. There's also a Download PDF button for sharing.
How do you verify an AI chat answer on the device detail page?
Click the citation bubbles to jump to the relevant highlight in the source document.
Reading rule for every project: how many summaries do you read in full?
At least the three most relevant 510(k) or De Novo summaries, in full. After that, use targeted chat questions to confirm your memory quickly. The tool supports this professional habit — it doesn't replace it.
5. Side-by-side comparison
Select multiple rows in the results table (aim for under ~10), then open the PDF Viewer tab. Ask one question — it goes to all selected devices in parallel, each with citations. This is the fastest way to compare and contrast devices: training data, PCCP scope, how they handled adding new scanners, and so on.
What does the side-by-side PDF viewer mode do?
Select multiple devices, open the PDF viewer tab, and ask one question (e.g., "Describe the training data"). It queries all selected devices simultaneously with citations, so you can compare and contrast quickly.
6. Collections
With rows selected, go to the Collections tab and create a labeled collection (e.g., "Cobb Angle Project"). Reload that selection any time — before a client call, pull up the collection and ask questions across all of its devices at once.
How do you save a set of selected devices for later use?
Select the rows, go to the Collections tab, and create a labeled collection (e.g., "Cobb Angle Project"). You can reload the selection anytime and carry it into the PDF viewer and other tabs that support selections.
7. Product codes and the regulations tree
Click a product code in the results to jump to it in the regulations tree — identification text, sibling product codes, and devices you can open in a PDF viewer on the right. Click a regulation number to see its identification, special controls, and related product codes. You can also search by product code or regulation number at the top of the tree. Always read the special controls if any exist for your device — it broadens your search and sharpens pre-kickoff research.
What can you do from the regulations tree view?
Browse product codes and regulation numbers, read the identification text and special controls, browse sibling product codes, open device PDFs on the right, and search by product code or regulation number at the top of the tree.
8. Chart view
Click Show Chart and segment by regulation number (or product code) to see which regulations dominate your result set. Clicking a regulation takes you into the regulations tree. Great for spotting that most matches are, say, hardware laparoscopic devices — a cue to go back and filter.
How do you see which regulations dominate a search result set?
Click "Show Chart" and segment by Regulation Number. Clicking a regulation takes you to the regulations tree.
9. The predicate graph
Open the Predicates tab for a family-tree view of predicate relationships. Click a node to trace its parents and children; selections from search carry over pre-selected. Commonly predicated devices are worth reading — a lot of people predicated them for a reason. The visual lineage is also handy on client calls, e.g. to show how a predicate family evolved and justify why your predicate still holds.
In the predicate graph, why are commonly predicated devices worth reading?
A lot of people predicated them for a reason. Clicking a node traces parents and children, and selections from search carry over pre-selected.
10. Embeddings: the galaxy map
The Embeddings tab plots every matching document in a 2-D "galaxy map" where semantically similar devices cluster together. Hover or click clusters to explore, and let AI label the clusters for you. Embeddings beat product codes for grouping: two devices can carry different product codes (LLZ vs. QIH) yet do the same thing — the embedding captures the meaning of the intended use and device story. This is also exactly how retrieval-augmented generation (RAG) works under the hood, and it makes a great visual on client calls.
Try it yourself
Head to the search page and work through a few of these AI/ML fuzzy searches to build intuition: perivascular fat on CT · aortic valve calcification opportunistic screening on noncontrast CT · breast cancer prediction on digital pathology slides · autism detection · gestational age prediction · a hearing aid that can also detect a pulse · foundation model based analysis of ECG · large language models · penetration test. Watch how the relevance scores, intended use, and AI Performance tables tell you when results stop being meaningful.