AUSTRAC’s compliance standards are rising, and Australian banks are responding by automating regulatory reporting to save time, improve accuracy, and strengthen trust.

Introduction

Regulatory reporting is one of the most critical pillars of Australia’s financial crime compliance framework. Every day, banks must identify, investigate, and report suspicious or high-value transactions to AUSTRAC within strict deadlines.

However, traditional reporting processes rely heavily on manual reviews and fragmented systems, which slow down investigations and increase the risk of human error. As transaction volumes surge through NPP and PayTo, the pressure on compliance teams has never been higher.

The solution lies in automation. By combining intelligent workflows, analytics, and AI, Australian banks can streamline their Suspicious Matter Reports (SMRs), Threshold Transaction Reports (TTRs), and International Funds Transfer Instructions (IFTIs), ensuring accuracy, speed, and regulatory confidence.

Why Regulatory Reporting Matters

Regulatory reporting allows AUSTRAC to identify and disrupt criminal activity such as money laundering, terrorism financing, and large-scale fraud. Each report contributes to Australia’s broader intelligence network, connecting financial data with law enforcement insights.

Failure to file complete or timely reports can lead to:

• Regulatory penalties: AUSTRAC can impose multi-million-dollar fines.
• Reputational damage: Missed or inaccurate reports undermine public trust.
• Operational inefficiencies: Manual processes consume valuable time and resources.

For banks, automation is not just a cost-saving measure. It is essential for maintaining compliance integrity in a real-time financial ecosystem.

Understanding the Three Core Reports

1. Suspicious Matter Reports (SMRs)

Banks must file SMRs when they detect behaviour that may involve money laundering, terrorism financing, tax evasion, or other financial crimes.

Key requirements:

• Must be lodged within 3 business days for suspected terrorism financing or within 10 business days for other suspicions.
• Must include transaction details, parties involved, and reasons for suspicion.
• Requires strong internal escalation procedures.

2. Threshold Transaction Reports (TTRs)

TTRs must be submitted for cash transactions of AUD 10,000 or more (or equivalent in foreign currency).

Key requirements:

• Must be filed within 10 business days.
• Applies to deposits and withdrawals.
• Includes both individuals and business customers.

3. International Funds Transfer Instructions (IFTIs)

IFTIs capture cross-border transactions entering or leaving Australia.

Key requirements:

• Must be reported within 10 business days.
• Applies to both electronic and physical fund transfers.
• Must include sender, receiver, amount, and intermediary details.

Challenges in Traditional Reporting Processes

1. Manual Data Gathering
   Analysts pull information from multiple systems, increasing the chance of incomplete or inconsistent data.
2. Delayed Investigations
   Manual review of alerts slows down the reporting cycle, risking regulatory breaches.
3. Human Error
   Copy-paste errors or missed fields can trigger report rejections or audits.
4. Disconnected Systems
   Transaction monitoring, case management, and reporting tools often operate in silos.
5. Growing Transaction Volumes
   With instant payments, even small institutions process millions of transactions daily.

These challenges make automation critical for accuracy, timeliness, and scalability.

Why Automation Is the Future of Compliance

1. Speed and Efficiency

Automated systems can prepare and submit SMRs, TTRs, and IFTIs within seconds, ensuring no deadlines are missed.

2. Accuracy and Consistency

AI-driven validation checks reduce data entry errors and enforce regulatory formatting standards automatically.

3. Real-Time Compliance

Automation allows continuous monitoring and instant report generation when suspicious activity is detected.

4. Audit-Ready Transparency

Every automated report includes full audit trails for easy review by compliance officers or regulators.

5. Reduced Workload

Analysts spend more time on complex investigations and less on administrative tasks.

6. Scalable for Growth

As institutions expand or integrate new payment channels, automated systems scale without proportional cost increases.

AUSTRAC’s View on Automation

AUSTRAC has consistently encouraged the adoption of RegTech and SupTech solutions that strengthen compliance. Automation aligns perfectly with AUSTRAC’s objectives:

• Ensuring timely and high-quality reporting.
• Reducing human error and operational delays.
• Encouraging explainable, risk-based use of AI.
• Promoting stronger collaboration between regulators and financial institutions.

Automated reporting systems must still include human oversight, ensuring that final reviews and validations remain accountable and transparent.

How Automated Reporting Works

1. Data Integration

Automation tools connect directly to core banking, transaction monitoring, and customer databases to capture relevant details in real time.

2. Alert Prioritisation

AI models score alerts based on risk level, automatically surfacing the most urgent or complex cases for review.

3. Case Preparation

Structured templates pre-populate SMRs, TTRs, and IFTIs with verified information, minimising manual input.

4. Workflow Automation

Reports move through approval stages automatically, with notifications for compliance officers and auditors.

5. Submission and Acknowledgement

Final reports are securely transmitted to AUSTRAC’s online reporting system, with confirmation logs stored for audit.

6. Continuous Improvement

Feedback from investigators refines model accuracy and improves future report quality.

Key Features of Automated Reporting Systems

1. End-to-End Integration: Connects data across AML, fraud, onboarding, and KYC systems.
2. Risk-Based Prioritisation: Automatically highlights transactions that match high-risk patterns.
3. Explainable AI Models: Every decision can be justified to regulators.
4. Federated Intelligence: Incorporates typologies shared through networks such as the AFC Ecosystem.
5. Audit Trail Management: Tracks every report creation, edit, and submission.
6. Real-Time Dashboards: Provide visibility into pending, completed, and overdue reports.

Case Example: Regional Australia Bank

Regional Australia Bank, a community-owned institution, demonstrates how automation can modernise compliance. By integrating advanced analytics and automated workflows, the bank has achieved faster SMR turnaround times and stronger data accuracy while maintaining transparency with AUSTRAC.

This success highlights how even mid-sized institutions can deliver world-class compliance through technology-driven efficiency.

Spotlight: Tookitaki’s FinCense

FinCense, Tookitaki’s AI-powered compliance platform, transforms regulatory reporting for Australian banks.

• Automated Reporting: Prepares and submits SMRs, TTRs, and IFTIs directly to AUSTRAC with zero manual intervention.
• Agentic AI Copilot (FinMate): Generates narrative summaries for each case, helping investigators finalise reports faster.
• Federated Intelligence: Accesses anonymised typologies from the AFC Ecosystem to strengthen detection accuracy.
• Integrated Case Management: Links alerts, investigations, and reports in a single interface.
• Explainable AI: Ensures transparency and auditability across all compliance processes.
• Real-Time Dashboards: Give compliance officers complete visibility into reporting pipelines.

With FinCense, banks can move from reactive compliance to predictive, automated reporting that keeps pace with AUSTRAC’s evolving requirements.

Benefits of Automating Regulatory Reporting

1. Zero Missed Deadlines: Reports are filed automatically as soon as suspicious activity is confirmed.
2. Fewer Human Errors: Automated validation ensures data integrity.
3. Faster Investigations: Case preparation time is reduced by up to 80 percent.
4. Better Audit Outcomes: Built-in transparency satisfies regulatory audits.
5. Enhanced Collaboration: Teams can focus on decision-making instead of data collection.
6. Improved Cost Efficiency: Reduced manual effort cuts operational costs while boosting compliance quality.

Implementation Roadmap for Banks

1. Assess Current Reporting Gaps: Identify inefficiencies in existing processes.
2. Integrate Systems: Connect AML, transaction monitoring, and KYC data through secure APIs.
3. Deploy Automation Tools: Choose platforms that meet AUSTRAC’s technical and regulatory requirements.
4. Train Compliance Teams: Ensure staff can review and validate automated outputs confidently.
5. Monitor Performance: Regularly review KPIs such as turnaround time, accuracy rate, and report volume.
6. Engage Regulators: Maintain open dialogue with AUSTRAC about system design and updates.

Future of Regulatory Reporting in Australia

1. AI-Assisted Case Writing: Intelligent copilots will generate entire SMRs automatically.
2. Real-Time Reporting: AUSTRAC may move toward live data feeds for continuous oversight.
3. Federated Learning Collaboration: Banks will share typologies securely without revealing customer data.
4. Predictive Compliance: Systems will identify suspicious activity before transactions settle.
5. Global Harmonisation: Automation will support compliance with both AUSTRAC and FATF standards.

Conclusion

Regulatory reporting is no longer a back-office function. It is a frontline defence that protects the integrity of Australia’s financial system.

Automation gives banks the agility to meet AUSTRAC’s demands without compromising accuracy or customer trust. Regional Australia Bank demonstrates how innovation, when combined with accountability, can set a new benchmark for compliance excellence.

With Tookitaki’s FinCense, Australian banks can automate reporting end to end, eliminate manual errors, and ensure continuous compliance in a world where speed and precision define success.

Pro tip: The best compliance systems do not just meet deadlines; they build trust by getting it right every single time.

In the Philippines, compliance teams are trading routine for intelligence — and Agentic AI is leading the charge.

The financial crime landscape in the Philippines has grown more complex than ever. From money mule networks and investment scams to online fraud syndicates, criminals are exploiting digital channels at unprecedented speed. Traditional compliance systems — reliant on static rules and manual reviews — are struggling to keep up.

This is where AML investigation software steps in. Powered by Agentic AI, these solutions are transforming how banks and fintechs detect, analyse, and respond to suspicious activity. In a region where regulatory scrutiny is tightening and financial innovation is accelerating, the Philippines stands at the front line of this transformation.

The Growing Burden on Compliance Teams

Financial institutions across the Philippines face increasing pressure to balance growth with risk management. The Anti-Money Laundering Council (AMLC) and the Bangko Sentral ng Pilipinas (BSP) have rolled out new regulations that demand stronger customer due diligence, more granular monitoring, and faster suspicious transaction reporting.

At the same time, the ecosystem has become more complex:

• Digital payment growth has created new entry points for fraud.
• Investment scams and online lending abuse continue to rise.
• Cross-border flows have made tracing illicit money trails harder.

These developments have turned compliance operations into a high-stakes race against time. Analysts often sift through thousands of alerts daily, many of which turn out to be false positives. What used to be an investigation problem is now an efficiency and accuracy problem — and the solution lies in intelligence, not just automation.

What AML Investigation Software Really Does

Modern AML investigation software isn’t just a case management tool. It’s a system designed to connect the dots across fragmented data, spot suspicious relationships, and guide investigators toward the right conclusions — faster.

Key Functions:

• Alert triage: Prioritising alerts based on risk, behaviour, and contextual intelligence.
• Entity resolution: Linking related accounts and transactions to reveal hidden networks.
• Case investigation: Collating customer data, transaction histories, and red flags into a single view.
• Workflow automation: Streamlining escalation, documentation, and reporting for regulatory compliance.

But the real leap forward comes with Agentic AI — a new generation of artificial intelligence that doesn’t just analyse data, but actively assists investigators in reasoning, decision-making, and collaboration.

Agentic AI: The New Brain Behind AML Investigations

Traditional AI systems rely on predefined rules and pattern matching. Agentic AI, on the other hand, is dynamic, goal-driven, and context-aware. It can reason through complex cases, adapt to new risks, and even communicate with investigators using natural language.

In AML investigations, this means:

• Adaptive Learning: The system refines its understanding with every case it processes.
• Natural Language Queries: Investigators can ask the system questions — “Show me all linked accounts with unusual foreign remittances” — and get instant, contextual insights.
• Proactive Suggestions: Instead of waiting for input, the AI can surface leads or inconsistencies based on evolving risk patterns.

For Philippine banks facing talent shortages and rising compliance workloads, this is a game changer. Agentic AI augments human intelligence — it doesn’t replace it — by taking on the repetitive tasks and surfacing what truly matters.

How Philippine Banks Are Embracing Intelligent Investigations

The Philippines’ financial sector is undergoing rapid digital transformation. With over 30% of adults now transacting through e-wallets, and a growing cross-border payments ecosystem, compliance complexity is only deepening.

Forward-looking banks and fintechs have begun integrating AML investigation software with Agentic AI capabilities to strengthen investigative accuracy and reduce turnaround times.

Adoption Drivers:

1. Regulatory alignment: AMLC’s focus on data-driven risk management is pushing institutions toward AI-enabled investigation workflows.
2. Operational efficiency: Reducing false positives and manual intervention helps cut compliance costs.
3. Fraud convergence: As fraud and AML risks increasingly overlap, unified intelligence is now essential.

Tookitaki has been at the forefront of this change — helping financial institutions in the Philippines and across ASEAN shift from rule-based monitoring to adaptive, intelligence-led investigation.

Key Features to Look for in AML Investigation Software

Choosing the right AML investigation software goes beyond automation. Financial institutions should look for capabilities that blend accuracy, explainability, and collaboration.

1. Agentic AI Copilot

A key differentiator is whether the software includes an AI copilot — an embedded assistant that interacts with investigators in real time. Tookitaki’s FinMate, for example, is a local LLM-powered Agentic AI copilot designed specifically for AML and fraud teams. It helps analysts interpret cases, summarise findings, and suggest next steps — all while maintaining full auditability.

2. Collaborative Intelligence

The most advanced platforms integrate collective intelligence from communities like the AFC Ecosystem, giving investigators access to thousands of real-world scenarios and typologies. This empowers teams to recognise emerging risks — from mule networks to crypto layering — before they spread.

3. Federated Learning for Data Privacy

In jurisdictions like the Philippines, where data privacy regulations are strict, federated learning enables model training without centralising sensitive data. Each institution contributes insights without sharing raw data — strengthening collective defence while maintaining compliance.

4. Explainability and Trust

Every AI-generated recommendation should be explainable. Systems like Tookitaki’s FinCense prioritise transparent AI, ensuring investigators can trace every output to its underlying data, model, and reasoning logic — critical for audit and regulator confidence.

5. Seamless Integration

Integration with transaction monitoring, name screening, and case management systems allows investigators to move from detection to disposition without losing context — an essential requirement for fast-moving compliance teams.

The Tookitaki Approach: Building the Trust Layer for Financial Crime Prevention

Tookitaki’s end-to-end compliance platform, FinCense, is designed to be the Trust Layer for financial institutions — combining collaborative intelligence, federated learning, and Agentic AI to make financial crime prevention smarter and more reliable.

Within FinCense, the FinMate AI Copilot acts as an investigation partner.

• It summarises alert histories and previous investigations.
• Provides contextual recommendations on next steps.
• Offers case narratives ready for internal and regulatory reporting.
• Learns from investigator feedback to continuously improve accuracy.

This human–AI collaboration is transforming investigation workflows. Philippine banks that once spent hours on case analysis now complete reviews in minutes, with greater precision and consistency.

Beyond efficiency, FinCense and FinMate align directly with the AMLC’s push toward explainable, risk-based approaches — helping compliance officers maintain trust with regulators, customers, and internal stakeholders.

Case Example: A Philippine Bank’s Digital Leap

A mid-sized bank in the Philippines, struggling with high alert volumes and limited investigation bandwidth, implemented Tookitaki’s AML investigation software as part of its broader FinCense deployment.

Within three months:

• False positives dropped by over 80%.
• Investigation time per case reduced by half.
• Analyst productivity improved by 60%.

What made the difference was FinMate’s Agentic AI capability. The system didn’t just flag suspicious behaviour — it contextualised each alert, grouped related cases, and generated draft narratives for investigator review. The outcome was faster resolution, better accuracy, and renewed confidence in the compliance function.

The Future of AML Investigations in the Philippines

The next phase of compliance transformation in the Philippines will be shaped by Agentic AI and collaborative ecosystems. Here’s what lies ahead:

1. Human-AI Co-investigation

Investigators will work alongside AI copilots that understand intent, interpret complex relationships, and recommend actions in natural language.

2. Continuous Learning from the Ecosystem

Through federated networks like the AFC Ecosystem, models will learn from typologies shared across borders, enabling local institutions to anticipate new threats.

3. Regulatory Collaboration

As regulators like the AMLC adopt more advanced supervisory tools, banks will need AI systems that can demonstrate traceability, explainability, and governance — all of which Agentic AI can deliver.

The result will be a compliance environment that’s not just reactive but predictive, where financial institutions detect risk before it manifests and collaborate to protect the integrity of the system.

Conclusion: Intelligence, Trust, and the Next Chapter of Compliance

The evolution of AML investigation software marks a turning point for financial institutions in the Philippines. What began as a push for automation is now a movement toward intelligence — led by Agentic AI, grounded in collaboration, and governed by trust.

As Tookitaki’s FinCense and FinMate demonstrate, the path forward isn’t about replacing human judgment but amplifying it with smarter, context-aware systems. The future of AML investigations will belong to those who can combine human insight with machine precision, building a compliance function that’s not only faster but fairer — and trusted by all.

As financial crime grows more complex, Australian banks are turning to AI and now Agentic AI to revolutionise how transactions are monitored and risks detected.

Introduction

Australia’s financial landscape is evolving fast. The growth of real-time payments, digital banking, and cross-border transactions has made detecting financial crime more challenging than ever. Traditional rule-based transaction monitoring systems, designed for slower and simpler payment environments, are no longer enough.

In response, Australian banks are increasingly adopting artificial intelligence (AI) to enhance the accuracy, speed, and adaptability of their AML programs. But the latest evolution, Agentic AI, is taking compliance to an entirely new level.

This blog explores how AI, and particularly Agentic AI, is transforming transaction monitoring across Australia’s banking sector, enabling faster detection, smarter investigations, and stronger regulatory alignment with AUSTRAC.

Why Transaction Monitoring Needs a New Approach

1. The Rise of Real-Time Payments

With the New Payments Platform (NPP) and PayTo, transactions clear in seconds. Fraudsters and launderers exploit this speed to move funds through multiple mule accounts before banks can react.

2. Sophisticated Criminal Tactics

Financial crime is no longer limited to simple structuring. Criminals use synthetic identities, cross-border layering, and digital assets to evade detection.

3. High False Positives

Rule-based systems trigger thousands of unnecessary alerts, overwhelming compliance teams and increasing costs.

4. AUSTRAC’s Evolving Standards

AUSTRAC expects continuous monitoring, explainability, and proactive detection. Banks must show they can identify suspicious activity before it spreads across the financial system.

5. Customer Experience Pressures

Delays or false flags impact legitimate customers. AI enables banks to balance security and service quality.

The Limitations of Traditional Monitoring

For years, transaction monitoring relied on static rules and thresholds — for example, flagging transactions over AUD 10,000 or rapid transfers to high-risk countries. While these methods catch known risks, they fail against sophisticated or adaptive schemes.

Limitations include:

• Static logic: Can’t detect new or subtle behaviours.
• Manual reviews: Investigators waste time on low-risk alerts.
• No learning loop: Systems don’t improve automatically over time.
• Fragmented data: Disconnected systems hinder visibility across channels.

In today’s fast-moving financial environment, static systems have become reactive rather than preventive.

How AI Transforms Transaction Monitoring

AI reshapes monitoring from a reactive process into a proactive intelligence system that continuously learns from data.

1. Machine Learning for Pattern Recognition

AI models analyse historical and real-time data to detect patterns that indicate suspicious activity — such as unusual fund flows, velocity changes, or repeated interactions with high-risk entities.

2. Behavioural Analytics

AI builds detailed customer profiles and detects deviations from normal behaviour, flagging potential risks that traditional systems miss.

3. Adaptive Thresholding

Instead of fixed thresholds, AI dynamically adjusts alert sensitivity based on risk context, reducing false positives.

4. Entity Resolution

AI connects fragmented data to identify relationships between customers, accounts, and devices — crucial for uncovering complex laundering networks.

5. Natural Language Processing (NLP)

AI interprets transaction narratives, case notes, and free-text fields, identifying hidden clues like invoice mismatches or unusual descriptions.

6. Continuous Learning

Every investigation outcome feeds back into the model, improving detection accuracy over time.

Agentic AI: The Next Frontier in Compliance

Agentic AI goes beyond traditional AI by combining autonomy, reasoning, and collaboration. Instead of just executing pre-trained models, Agentic AI acts as an intelligent assistant that can:

• Analyse transactions and contextual data.
• Generate risk summaries in natural language.
• Recommend actions based on regulatory frameworks.
• Learn from investigator feedback to improve continuously.

In compliance, this means faster decisions, fewer manual errors, and higher operational efficiency.

How Agentic AI Works in Transaction Monitoring

1. Data Ingestion and Contextual Understanding

Agentic AI continuously consumes structured (transactions, KYC) and unstructured (case notes, communications) data to form a full risk picture.

2. Dynamic Risk Scoring

It assigns real-time risk scores to each transaction, considering behavioural patterns, customer history, and contextual anomalies.

3. Intelligent Narration

When a transaction is flagged, Agentic AI can summarise the alert — describing what happened, why it matters, and what actions are recommended — in clear, regulator-friendly language.

4. Self-Learning Capabilities

Each closed case improves its reasoning. Over time, the system develops institutional knowledge, adapting to new typologies without reprogramming.

5. Investigator Collaboration

Acting as a compliance copilot, Agentic AI assists investigators in triaging alerts, finding linked accounts, and preparing Suspicious Matter Reports (SMRs).

Benefits of AI and Agentic AI for Australian Banks

1. Significant False Positive Reduction: AI models prioritise relevant alerts, cutting investigation workload by up to 90 percent.
2. Improved Accuracy: Continuous learning enhances detection of new typologies.
3. Faster Investigations: Agentic AI copilots summarise and contextualise alerts in seconds.
4. Regulatory Confidence: Explainable AI ensures transparency and auditability for AUSTRAC.
5. Enhanced Customer Trust: Real-time, intelligent monitoring prevents fraud without disrupting legitimate transactions.
6. Operational Efficiency: Reduced manual workload lowers compliance costs.

AUSTRAC’s View on AI in Compliance

AUSTRAC has encouraged innovation in RegTech and SupTech solutions that enhance financial integrity. Under the AML/CTF Act, AI-powered systems are acceptable if they:

• Maintain auditability and explainability.
• Apply risk-based controls.
• Support timely and accurate reporting.
• Are regularly validated and reviewed for bias and accuracy.

AUSTRAC’s collaboration with technology providers reflects a growing recognition that AI is essential to managing modern financial crime risks.

Case Example: Regional Australia Bank

Regional Australia Bank, a community-owned institution, has embraced AI-driven compliance to enhance its transaction monitoring capabilities. By leveraging intelligent analytics, the bank has reduced investigation time, improved accuracy, and strengthened its reporting processes — all while maintaining customer trust and transparency.

Its experience demonstrates that AI adoption is not limited to large institutions; even mid-sized banks can lead in compliance innovation.

Spotlight: Tookitaki’s FinCense and Agentic AI

FinCense, Tookitaki’s flagship compliance platform, integrates Agentic AI to redefine transaction monitoring for Australian banks.

• Real-Time Monitoring: Analyses millions of transactions across NPP, PayTo, and international payments instantly.
• Agentic AI Copilot (FinMate): Assists investigators by narrating alerts, identifying linked parties, and generating regulatory summaries.
• Federated Intelligence: Utilises anonymised typologies contributed by the AFC Ecosystem to detect new risks collaboratively.
• Explainable AI: Ensures every model decision is transparent, auditable, and regulator-ready.
• End-to-End Case Management: Combines fraud, AML, and sanctions monitoring into a unified workflow.
• AUSTRAC Alignment: Automates SMRs, TTRs, and IFTIs with full compliance assurance.

With Agentic AI at its core, FinCense transforms transaction monitoring from a static process into an intelligent, adaptive system that anticipates risk before it happens.

Implementing AI-Driven Monitoring: Best Practices

1. Start with Clean Data: High-quality data ensures reliable model performance.
2. Adopt Explainable Models: Regulators prioritise transparency in AI decision-making.
3. Integrate AML and Fraud Operations: Unified systems enhance efficiency.
4. Invest in Investigator Training: Equip teams to work alongside AI tools effectively.
5. Validate Models Regularly: Continuous testing maintains fairness and accuracy.
6. Collaborate through Federated Intelligence: Shared insights strengthen detection across institutions.

Future of Transaction Monitoring in Australia

1. Predictive Compliance: Systems will forecast risks and block suspicious transactions before they occur.
2. Hyper-Personalised Risk Scoring: AI will assess risk at the individual customer level in real time.
3. Industry-Wide Collaboration: Federated learning will connect banks for collective intelligence.
4. Agentic AI Investigators: Autonomous copilots will handle tier-one alerts end to end.
5. RegTech-Regulator Integration: AUSTRAC will increasingly rely on direct system data feeds for oversight.

Conclusion

The future of transaction monitoring in Australia lies in intelligence, not volume.
AI enables banks to uncover complex, hidden risks that traditional systems miss, while Agentic AI brings a new level of automation, reasoning, and transparency to compliance operations.

Regional Australia Bank shows that innovation is achievable at any scale. With Tookitaki’s FinCense and its built-in Agentic AI, Australian banks can move beyond reactive monitoring to real-time, proactive financial crime prevention — strengthening both compliance and customer trust.

Pro tip: The smartest transaction monitoring systems don’t just detect suspicious activity; they understand it, explain it, and learn from it.