The GenAI augmented actuary: Crossing the inflection point 

The actuarial profession, one of the oldest and most respected in the financial world, finds itself at an inflection point as it grapples with the rapid advancement of generative artificial intelligence (GenAI) technologies.  

On one hand, GenAI has the potential to transform the actuarial landscape, serving as a powerful tool to augment the actuarial skillset. On the other hand, the rapid pace of GenAI development creates a risk that actuaries will fall behind in this technological revolution. As Richard Baldwin, a renowned economist, warned at the World Economic Forum in 2022, "AI won't take your job. It's somebody using AI that will take your job."

Actuaries must upskill to stay relevant and useful in this increasingly tech-driven landscape. By embracing the transformative power of GenAI, they can reposition themselves as leaders in using data and quantitative techniques to inform judgment-based decisions and cement their indispensable role in the industries they serve. 

To bring this vision to reality, actuaries, their leaders, and the institutions that support them, must understand and embrace what GenAI adoption means for the actuarial profession and the important work they do. It’s time to acknowledge the significant change ahead and mobilise to take the actions necessary for the profession to reap the benefits that GenAI promises. 

Actuaries are bound by a set of professional obligations that govern their work, ensuring accountability and responsibility, and maintaining the reputation of the profession. These obligations extend to the use of GenAI by actuaries which creates specific implications that may not be present in other professions. As the actuarial profession embraces the transformative potential of GenAI, it is crucial to consider the interplay between the profession's stringent standards, ethical codes, and regulatory requirements and these advanced technologies. 

Professional requirements 

At the core of the actuarial profession is a unique proposition that differentiates actuaries from the likes of data scientists and machine learning engineers: a commitment to professionalism, adherence to a code of conduct, ethical principles, and domain-specific knowledge often gained through the exam system and many years of apprenticeship-based rotations. Actuaries around the world are bound by rigorous standards of practice that require them to consider the appropriateness and limitations of their analyses. They must be able to explain and defend their methodologies, assumptions, and conclusions in a manner that is appropriate and comprehensible to their audience. Furthermore, actuaries are expected to exercise professional judgment and adhere to ethical principles, such as acting with integrity, objectivity, and competence. These standards and professional obligations still apply as GenAI tools and solutions become increasingly prevalent in the delivery of actuarial services.  

With GenAI technologies rapidly evolving, actuarial standards and professional bodies will need to quickly adapt and evolve to provide guidance on the responsible and ethical use of these tools. This could involve developing new standards or guidelines specifically addressing the integration of GenAI into actuarial practices. Professional standards that codify best practices around ethical AI development, model validation and governance will be critical, but getting them in place quickly will be important to keeping the profession relevant. With such standards in place, actuaries can lead the way by leveraging their risk management expertise to ensure AI is adopted responsibly, building trust while unlocking its potential benefits. The leadership of the actuarial community will be pivotal to realizing the promise of GenAI while mitigating individual, business and societal risks.

Regulatory considerations

The regulatory landscape for GenAI models is still evolving and varies across different jurisdictions and domains. However, common themes and principles, such as data protection, privacy, ethics, transparency, and accountability can be identified. These principles aim to ensure that models are used in a way that respects individuals rights, and that they do not cause harm, discrimination, or deception to individuals or society. The careful balance between AI capability and regulation is a fine line that will play out in different regions in the next few years. 

The European Union’s proposed legal framework for AI classifies GenAI models into various risk-based categories. The framework establishes a risk-based approach, where the level of regulation depends on the potential impact of the AI system on human health, safety, and fundamental rights. The framework also introduces sanctions and remedies for non-compliance, as well as a system of monitoring and enforcement. 

The United States, on the other hand, has adopted a more flexible and sector-specific approach, relying on, or expanding, existing laws and regulations to address the issues raised by GenAI models. The approach emphasises the role of self-regulation, voluntary standards, and best practices, as well as the promotion of innovation and competitiveness. It recognises the need for coordination and collaboration among federal agencies, state and local governments, and international partners.  

In addition, there are various industry-specific and professional standards and guidelines that apply to GenAI, such as the ASOP 56 on Modeling (US), TAS 100 model guidance (UK) and APN 901 (ZA) which mandates that models are used for their intended purpose and requires model understandability and validation procedures. These standards and guidelines, along with others, provide guidance and recommendations for actuaries on how to use GenAI and large language models (LLMs) in a manner that is consistent with their professional obligations and values, as well as the expectations and interests of their clients, employers, regulators, and the public. 

Implications for actuaries 

Given the professional obligations and regulations that govern actuarial work, some implications particularly critical to consider include: 

• Actuaries will need to maintain their knowledge of relevant regulations and standards that apply to them, as well as update their knowledge and skills on GenAI. This will require continuous learning, training, and certification. Actuaries will also need to stay informed of the latest GenAI developments and trends, as well as the potential benefits and risks they pose for their work and the industries they serve. This is an opportunity both for the professional bodies and employers to provide timely and relevant upskilling for actuaries. 
• Consistent with what makes the profession unique, actuaries will need to ensure that their use is ethical, transparent, and accountable, and that they adhere to the principles of professionalism, integrity, and public interest. This will require clear documentation, communication, and disclosure of the methods, assumptions, limitations, and uncertainties of GenAI. Actuaries will also need to justify and explain their use of GenAI to their clients, employers, regulators, and the public, and to address any concerns that may arise. 
• Actuaries will need to assess and manage the risks and opportunities associated with GenAI, such as data quality, bias, fairness, accuracy, reliability, security, and liability.  
• Actuaries will also need to monitor and evaluate the performance and impact of GenAI, and to report and correct any errors, failures, or incorrect outputs. This should be embedded into the monitoring that is required in any process through the lens of the actuarial control cycle.

From data processing and model building to risk assessment and governance, AI presents opportunities for actuaries to harness its potential while upholding professional standards and ethical principles. By embracing machine learning, natural language processing, and deep learning capabilities, actuaries can streamline routine tasks, uncover intricate patterns within complex data sets, and develop more accurate and sophisticated financial/risk models.  

GenAI, specifically, can automate many of the routine tasks that actuaries perform, such as report writing, data analysis, model development and certain aspects of communication. These tasks often consume a significant amount of time and resources and can be prone to human errors and biases. By using GenAI, actuaries can delegate analytical and basic writing tasks to intelligent systems that can produce high-quality outputs for actuarial review faster and more reliably. This can free up time for actuaries to focus on more productive areas such as strategic initiatives, results interpretation, value-adding analysis, complex communications and decision-making.  

GenAI can also enhance and extend traditional actuarial methods, improving the accuracy and sophistication of the approaches. Often challenges with data quality, availability, and relevance can limit the scope and validity of actuarial methods. GenAI can help actuaries increase the robustness and reliability of models by augmenting data sets with synthetic data that preserves the statistical properties and correlations of the original data, while generating new and realistic observations. 

Actuaries can also use GenAI to generate realistic and diverse scenarios for risk modelling that take into account the uncertainty and complexity of the real world, including impacts from factors such as market movements, pandemics, and regulatory changes. It can help actuaries understand and address non-modelled, secondary peril and emerging risks, such as climate risk, product features of Specie insurance coverage (fine art, NFTs etc.), and niche exceptions and endorsements for cyber risks.  

GenAI can also lower the barriers to entry for actuarial professionals to explore wider fields, such as finance, energy, and technology, as it reduces the need for specialised tools and steep learning curves. 

Consider, for example, these seven GenAI actuary use cases:

In addition to GenAI bringing many benefits, the integration of AI into actuarial practices raises critical considerations around ensuring transparency, fairness and explainability, addressing potential biases, and safeguarding data privacy and security. As AI continues to evolve, actuaries have a pivotal role in shaping its responsible adoption, combining risk expertise with AI capabilities to unlock new frontiers in assessment, product innovation, and decision-making while adapting skill sets to effectively design, validate, and govern AI systems. To take full advantage of what GenAI can offer the profession, actuaries will need to recognise the tools’ challenges and limitations and develop strategies to combat them.

Model transparency, explainability, data quality and bias

GenAI can introduce many of the same concerns as more traditional AI, and the
expectation of Responsible AI use is fully applicable (see PwC’s view on Responsible AI). For example, actuaries must ensure they combat the “black box” inability to explain models as well as the bias that can be introduced through training data. GenAI can also hallucinate, or produce information, images, or text that are not based on factual data or reality. Actuaries will need to assess and manage risks such as data quality, bias, fairness, accuracy and reliability. This will require robust testing, validation, and verification of GenAI and LLMs as well as the implementation of appropriate controls, safeguards, and mitigation measures.

Data privacy

Another critical concern with GenAI is the ethical and legal implications of data privacy and ownership. Actuaries often work with sensitive and personal data, and the use of GenAI necessitates strict adherence to data privacy regulations. Ensuring that data used in AI models is anonymised, and that its ownership is clearly defined, is paramount to avoiding legal pitfalls and maintaining client trust.

Fears of obsolescence and need for skills enhancement

On the human side, many professionals fear the potential of being "replaced" by AI, since it can perform tasks that those professionals currently handle. It is important for teams to understand that GenAI does not replace the need for actuaries. In fact, the skills of the actuary have never been more important. While AI decisions are based solely on technical factors, the actuary can bring our professionalism requirements and standards of practice, ensuring a more comprehensive and well-rounded decision-making process. GenAI does, however, bring with it a critical need to enhance actuaries’ skills to effectively manage and utilise AI in a responsible manner. Through proper training, actuaries can emerge as a crucial contributor to the decision-making process.

To thrive in the future world of AI, a cultural shift in the actuarial profession is needed— one in which we collectively embrace GenAI, embed it into the actuarial education and skillset, invest in its research and development for actuarial-specific use cases, and accelerate the transition to a future in which GenAI is part of the profession’s DNA.

Actuarial professional bodies worldwide must become the nexus behind this cultural shift and accelerate the development of guidelines and best practices to keep pace with technological progress. They must take proactive steps to ensure that actuaries are well-equipped to harness the power of GenAI while preserving the integrity and reputation of the profession.

For new actuaries, it is essential to push the relevance of what is covered in the education system. The current actuarial syllabus will always struggle to keep up with the pace of technology change as we are seeing with GenAI, leaving many professionals underprepared for integrating these technologies into their work. Finding ways to continuously update the syllabus to include comprehensive AI training will ensure that future actuaries are not only literate in the world of GenAI but can also drive the future of the profession.

For current actuaries, upskilling is imperative. Training programs should cover GenAI and its underlying emerging technologies. Actuaries must adapt rapidly to remain relevant and to be at the forefront of using GenAI to ensure professional competence and judgment is applied. Even senior actuaries who no longer engage in routine actuarial tasks and analytical work, and instead focus on expert judgment and decision-making, should prioritise upskilling in GenAI. This knowledge will be crucial to ensure the thought process behind their decisions and recommendations are underpinned by accurate knowledge of how GenAI was used in the work.

The profession must also actively engage in the research and development of GenAI to create useful but transparent and interpretable tools. Actuarial leadership must be proactive in spearheading the creation of risk frameworks and governance principles related to the responsible use of GenAI in actuarial work and broader risk management.

Finally, as in many professions, legacy technology and data often restrict what actuaries can do. Companies will need to modernise their data and technology platforms to truly capitalise on the power of GenAI and enable the revolutionary new capabilities it promises.

As GenAI technologies rapidly advance, the actuarial profession is indeed at an inflection point—one in which, ultimately, GenAI will be the day-to-day tool to augment actuarial work and be the catalyst for a new era in the profession. To remain relevant and uphold ethical standards, the industry’s professional and educational institutions, organisations and actuaries need to work together to embrace GenAI and accelerate the transition to this new era. Here are actions each constituent can take today. 

Professional bodies

• Enhance the actuarial syllabus to incorporate relevant and up-to-date training in GenAI principles, concepts, development and uses, including ethical considerations.
• Update exam systems and professional development programs to help actuaries gain domain-specific knowledge and experience beyond the entry level tasks that GenAI can, and increasingly will, do efficiently. 

Actuarial leaders

• Provide introductory information to actuaries and company leadership so there is a common understanding of what GenAI is, the opportunities it presents for actuaries and the challenges that the team will need to stay cognisant of as it begins the GenAI journey.
• Form a small team charged with developing and prioritising GenAI use cases focused both on revenue and productivity enhancements.
• Actively engage with other parts of the company to execute pilots of priority use cases.

Individual actuaries

• Proactively engage with professional bodies, regulators and communities of interest to help shape the future of the GenAI-enabled actuarial profession.

As the profession looks ahead at the rapid advancement of GenAI technologies, it is an exciting time to be an actuary. Those on the forefront stand to gain from both the benefits GenAI brings to the actuarial role and the leadership that GenAI-enabled actuaries can bring to their organisations, their industries, and to society.

^{[1] PwC's Global Actuarial Modernisation Survey, 2023}