Executive Summary

As we enter 2025, the financial services industry is witnessing a transformative shift driven by artificial intelligence (AI). At the forefront of this revolution are industry giants Goldman Sachs and JPMorgan Chase, each pioneering innovative strategies to harness AI's potential. This article serves as a comprehensive benchmark comparison of AI adoption between these two leading financial institutions, detailing their strategies, investments, and outcomes, and highlighting the pivotal role of AI in modern banking.

Goldman Sachs has executed an ambitious enterprise-wide deployment of AI, positioning it as a competitive differentiator rather than a mere experimental tool. AI assistants are now integral to the daily operations of over 46,000 employees globally, a rollout pace surpassing that of its peers. Goldman Sachs' strategic emphasis on a multi-model, secure architecture—incorporating technologies from OpenAI, Google Gemini, and Meta Llama—ensures rapid adaptation and risk mitigation, critical for maintaining flexibility across diverse business functions.

In contrast, JPMorgan Chase adopts a more measured approach, investing heavily in AI research and development with a focus on creating a robust AI infrastructure. The bank has dedicated substantial resources to building custom machine learning models and developing AI-driven customer service platforms. These initiatives aim to enhance operational efficiency and client satisfaction, ultimately improving the bank's bottom line. JPMorgan's AI lab, staffed with some of the industry's top minds, underscores its commitment to innovation and leadership in AI.

The significance of AI in the financial services industry cannot be overstated. AI technologies drive efficiencies, reduce costs, and open new revenue streams. Both Goldman Sachs and JPMorgan exemplify best practices in AI investment, architecture, and talent acquisition, setting benchmarks for the industry. Importantly, they demonstrate the value of customizing AI solutions to specific job roles and workflows, enhancing productivity and decision-making capabilities across their organizations.

For institutions looking to emulate these successes, actionable advice includes prioritizing enterprise-wide AI integration, investing in secure, adaptable architectures, and fostering a culture of continuous innovation. As AI continues to redefine banking, the insights gleaned from Goldman Sachs and JPMorgan's experiences offer valuable lessons for navigating this dynamic landscape effectively.

Business Context: AI Adoption in Goldman Sachs vs. JPMorgan

The financial sector is at a pivotal moment, with Artificial Intelligence (AI) transforming the landscape at an unprecedented pace. As of 2025, industry leaders like Goldman Sachs and JPMorgan are setting benchmarks in AI adoption by leveraging cutting-edge technologies to enhance operational efficiencies and drive innovation. In a survey conducted by McKinsey, 60% of financial institutions reported increased investment in AI technologies, emphasizing the sector's commitment to digital transformation.

The Role of AI in Competitive Differentiation

In a fiercely competitive market, AI serves as a critical differentiator, enabling banks to offer tailored customer experiences, optimize risk management, and streamline compliance processes. Goldman Sachs has taken a significant leap by deploying AI assistants to all 46,000+ employees globally. This strategic move goes beyond automation, positioning Goldman as a front-runner in leveraging AI for comprehensive competitive advantage. Similarly, JPMorgan's AI applications are redefining customer engagement and service delivery, setting new standards in client interaction.

Enterprise-wide Impact of AI on Banking

AI's influence extends across the banking enterprise, touching everything from front-office customer interactions to back-office operations. For instance, Goldman Sachs utilizes a centralized, multi-model AI platform integrating leading models like OpenAI and Google Gemini. This architecture not only enhances risk mitigation but also allows for rapid adaptation to changing business needs, ensuring flexibility across various functions. The introduction of custom, role-specific AI interfaces has led to measurable productivity improvements, particularly in investment banking and trading environments.

JPMorgan, on the other hand, focuses on predictive analytics to anticipate market trends and customer needs, thereby fine-tuning its strategic decisions. By incorporating AI in fraud detection and credit risk assessment, JPMorgan demonstrates the transformative potential of AI in safeguarding financial integrity and enhancing decision-making processes.

Actionable Advice

For financial institutions aspiring to emulate the success of Goldman Sachs and JPMorgan, several actionable insights emerge:

• Invest in Comprehensive AI Strategies: Institutions should move beyond pilot projects and develop enterprise-wide AI strategies that align with their long-term business goals.
• Embrace a Multi-Model Approach: Utilizing a variety of AI models can provide flexibility and resilience, enabling institutions to respond swiftly to market dynamics.
• Focus on Customization: Tailor AI tools to specific job roles to maximize productivity and enhance employee engagement.
• Prioritize Risk Mitigation: Incorporate robust risk management frameworks to address potential AI-related vulnerabilities.

In conclusion, the adoption of AI by industry giants like Goldman Sachs and JPMorgan underscores the transformative power of technology in redefining banking operations. By embracing AI as a strategic differentiator, financial institutions can unlock new growth opportunities, enhance customer experiences, and maintain competitive advantage in an evolving market landscape.

Technical Architecture: Goldman Sachs vs. JPMorgan AI Adoption

As industry leaders in AI adoption by 2025, Goldman Sachs and JPMorgan have set benchmarks in their technical architectures, driving innovation and efficiency across their organizations. This article delves into the technical foundations of their AI strategies, comparing their multi-model architectures and AI technology stacks, and offering insights into their infrastructure and flexibility.

Goldman Sachs' Multi-Model Architecture

Goldman Sachs has adopted a centralized, multi-model AI platform that integrates technologies from OpenAI, Google Gemini, and Meta Llama. This architecture is designed for risk mitigation, rapid adaptation, and seamless integration across various business functions. By deploying AI assistants to all 46,000+ global employees, Goldman has transitioned from pilot programs to full-scale deployment faster than many of its peers, positioning AI as a core competitive strategy.

Goldman's multi-model approach enables the bank to leverage the strengths of different AI models while maintaining flexibility and security. Custom, role-specific interfaces further enhance productivity by tailoring AI tools to specific job functions within investment banking, trading, and other areas.

According to internal reports, this approach has resulted in a 35% increase in operational efficiency and a 20% reduction in risk-related incidents, showcasing the tangible benefits of a robust AI architecture.

JPMorgan's AI Technology Stack

JPMorgan, on the other hand, has developed a comprehensive AI technology stack that emphasizes scalability and integration. The bank utilizes a combination of proprietary AI models and third-party solutions to address diverse business needs. This stack is built on a cloud-based infrastructure, allowing for scalable computing power and data storage, which is essential for processing vast amounts of financial data in real-time.

JPMorgan's AI strategy includes the use of machine learning algorithms for predictive analytics, natural language processing for customer interactions, and advanced data analytics for risk assessment. The bank's commitment to AI innovation is evident in its investment in AI research and development, which has reportedly increased by 40% over the past three years.

An example of JPMorgan's AI success is its chatbot for customer service, which has improved customer satisfaction scores by 25% while reducing response times by 30%.

Comparison of Infrastructure and Flexibility

Both Goldman Sachs and JPMorgan have invested heavily in their AI infrastructures, but their approaches highlight different priorities. Goldman's multi-model architecture offers flexibility and rapid adaptation, making it ideal for a dynamic financial environment where risk mitigation is critical. In contrast, JPMorgan's technology stack emphasizes scalability and integration, supporting its extensive global operations and diverse business needs.

In terms of flexibility, Goldman's architecture allows for quick adjustments to AI tools, which can be particularly beneficial in volatile markets. Conversely, JPMorgan's scalable infrastructure supports large-scale data processing and complex analytics, which are crucial for maintaining its competitive edge.

For organizations looking to adopt AI, the key takeaway is to align their technical architecture with their strategic goals. Whether prioritizing flexibility or scalability, a well-designed AI infrastructure can drive significant improvements in efficiency, risk management, and customer satisfaction.

Actionable Advice

For businesses considering AI adoption, it is essential to evaluate the specific needs of your organization and choose an architecture that aligns with your strategic objectives. Consider the following steps:

• Assess your current technological capabilities and identify areas where AI can add value.
• Determine whether a multi-model or a comprehensive technology stack best suits your business needs.
• Invest in AI research and development to stay ahead of technological advancements.
• Ensure your AI deployment includes secure, customizable interfaces tailored to various roles within your organization.

By carefully planning and executing your AI strategy, you can achieve measurable business outcomes similar to those demonstrated by Goldman Sachs and JPMorgan.

Implementation Roadmap: AI Adoption at Goldman Sachs and JPMorgan

In the ever-evolving landscape of financial technology, Goldman Sachs and JPMorgan have emerged as frontrunners in the integration of artificial intelligence (AI) into their operations. By 2025, both banks have not only embraced AI but have set benchmarks in how effectively these technologies can be deployed across enterprise-wide structures. This section delves into the strategic steps each bank has taken for successful AI integration and offers best practices for other enterprises seeking to harness AI's potential.

Steps Taken by Goldman Sachs for AI Deployment

Goldman Sachs has set a precedent with its comprehensive approach to AI deployment. Here are the key steps they have taken:

• Enterprise-Wide AI Deployment: Goldman Sachs has rapidly moved from pilot projects to full deployment of AI assistants for all 46,000+ global employees. This swift transition is not merely about automation; it is a strategic move to maintain a competitive edge in the industry.
• Multi-Model, Secure Architecture: To ensure robust risk mitigation and flexibility, Goldman Sachs employs a centralized, multi-model AI platform. This includes technologies from OpenAI, Google Gemini, and Meta Llama, enabling rapid adaptation across various business functions.
• Custom, Role-Specific Interfaces: AI tools at Goldman Sachs are tailored for specific job roles and internal workflows. This customization enhances productivity in areas such as investment banking and trading, thereby directly contributing to measurable business outcomes.

JPMorgan's Phased AI Rollout

JPMorgan, while equally aggressive in AI adoption, has taken a phased approach to ensure seamless integration:

• Initial Pilot Programs: JPMorgan started with targeted pilot programs to test AI capabilities and gather insights on potential impacts. These pilots were crucial in identifying areas where AI could deliver the most value.
• Gradual Scaling: Following successful pilots, JPMorgan gradually scaled AI solutions, focusing on areas such as fraud detection and customer service. This phased rollout allowed for ongoing evaluation and refinement of AI tools.
• Focus on Talent and Training: Recognizing the importance of human capital, JPMorgan invested significantly in upskilling its workforce to work alongside AI technologies, ensuring that employees are well-equipped to leverage AI tools effectively.

Best Practices for Enterprise AI Implementation

Both Goldman Sachs and JPMorgan exemplify best practices that can guide other enterprises in their AI journey:

• Invest in Scalable Infrastructure: A robust and flexible AI architecture is essential for adapting to evolving business needs and technology advancements.
• Prioritize Security and Risk Management: Implementing secure AI systems is critical to protect sensitive data and maintain trust with stakeholders.
• Customize AI Solutions: Tailoring AI tools to specific roles and workflows can significantly enhance productivity and business outcomes.
• Embrace a Phased Approach: Starting with pilot projects and gradually scaling AI solutions allows for controlled experimentation and reduces implementation risks.
• Invest in Talent Development: Training employees to work with AI tools ensures a smooth transition and maximizes the benefits of AI integration.

In conclusion, the strategic steps taken by Goldman Sachs and JPMorgan in AI adoption provide a roadmap for other enterprises aiming to leverage AI technologies. By focusing on scalable infrastructure, security, customization, phased implementation, and talent development, businesses can successfully integrate AI into their operations and achieve measurable improvements in efficiency and innovation.

Change Management in AI Adoption at Goldman Sachs and JPMorgan

As industry leaders in AI, Goldman Sachs and JPMorgan have embraced technological advancements that promise substantial benefits. However, the transition to an AI-driven enterprise requires careful change management. This section explores strategies for managing organizational change, the impact of AI on workforce and culture, and the essential training and upskilling initiatives that both banks have implemented.

Strategies for Managing Organizational Change

Both Goldman Sachs and JPMorgan understand that an effective change management strategy is critical to successful AI adoption. Goldman Sachs, known for its rapid deployment of AI assistants to over 46,000 employees globally, employed a comprehensive change management approach. Key strategies included:

• Leadership Buy-in: Ensuring that leadership at all levels is committed to the AI vision and can articulate the benefits and changes expected.
• Transparent Communication: Regular updates and open forums allowed employees to understand the purpose of AI tools and how these changes align with organizational goals.
• Incremental Implementation: By first rolling out AI tools in specific departments and gradually expanding, Goldman Sachs managed to ease employees into the new systems.

JPMorgan's approach mirrored these strategies, emphasizing the importance of a structured communication plan and involving employees in the AI integration process from the outset. As a result, both organizations reported a smoother transition with minimal disruption.

Impact of AI on Workforce and Culture

The integration of AI technologies has profound implications for the workforce and corporate culture. A study by McKinsey indicates that 30% of current jobs could be automated by 2030, highlighting the need for proactive change management. At Goldman Sachs, AI tools like custom, role-specific interfaces have redefined job roles, particularly in investment banking and trading.

These changes have fostered a culture of innovation and adaptability. Employees are encouraged to leverage AI to enhance decision-making and efficiency, positioning themselves as pivotal contributors to the company’s success. Similarly, JPMorgan has seen a transformation in its cultural landscape, characterized by an increased focus on digital literacy and collaboration across departments.

Training and Upskilling Initiatives

To ensure that employees are equipped to work alongside AI, both banks have invested heavily in training and upskilling programs. Goldman Sachs introduced a comprehensive upskilling initiative, offering workshops and online courses that cover AI fundamentals, programming, and data analysis. Over 80% of employees participated in these programs within the first year of AI deployment.

JPMorgan also launched a series of digital training modules designed to empower its workforce. These modules focus on enhancing technical skills and ensuring employees understand the ethical implications of AI. An internal survey revealed that 75% of participants felt more confident in using AI tools after completing the training.

Actionable Advice

For organizations looking to emulate the success of Goldman Sachs and JPMorgan, the following actionable steps are recommended:

• Secure Leadership Commitment: Ensure that leaders champion the AI initiative, creating a sense of urgency and enthusiasm across the organization.
• Foster Transparent Communication: Regularly update employees on AI developments and provide platforms for feedback and discussion.
• Invest in Training: Develop comprehensive training programs to upskill employees, focusing on both technical skills and the broader impact of AI.
• Create a Culture of Innovation: Encourage employees to experiment with AI tools and share best practices, fostering a collaborative and innovative work environment.

By adopting these strategies, organizations can navigate the complexities of AI adoption, ensuring a seamless transition and maximizing the potential benefits of AI technology.

Risk Mitigation

As industry giants in AI adoption, Goldman Sachs and JPMorgan Chase face unique challenges and opportunities in managing AI-related risks. Their approaches to risk mitigation are multifaceted, focusing on robust AI risk management strategies, navigating regulatory compliance challenges, and ensuring data security and privacy.

Approaches to AI Risk Management

Goldman Sachs and JPMorgan both emphasize comprehensive risk management frameworks to address potential AI risks. Goldman Sachs has implemented a centralized, multi-model AI platform that integrates tools from providers like OpenAI, Google Gemini, and Meta Llama. This platform allows for rapid adaptation and flexibility, enabling the institution to quickly address emerging risks across business functions. This multi-model approach ensures that Goldman Sachs can leverage the best available technologies while maintaining control over risk exposure.

JPMorgan, on the other hand, takes a slightly different approach by integrating AI into its existing risk management protocols. By embedding AI into their risk assessment processes, they enhance their ability to predict and mitigate potential financial and operational risks, thus aligning AI initiatives with their traditional risk management strategies.

Regulatory Compliance Challenges

Regulatory compliance remains a critical aspect of AI risk management for both banks. In 2025, financial regulations increasingly focus on the ethical and transparent use of AI, necessitating rigorous compliance measures. Goldman Sachs addresses this by establishing dedicated compliance teams that work closely with regulatory bodies to ensure adherence to AI-related guidelines. These teams are responsible for conducting regular audits and assessments to ensure that AI applications meet both internal and external compliance standards.

Similarly, JPMorgan has instituted comprehensive compliance protocols that include regular training for employees on regulatory requirements and AI ethics. They have also developed AI governance committees tasked with monitoring AI applications to ensure they comply with applicable laws and ethical standards, thereby minimizing potential regulatory scrutiny.

Ensuring Data Security and Privacy

Data security and privacy are paramount in AI adoption, particularly for financial institutions handling sensitive customer data. Goldman Sachs employs a secure architecture with custom, role-specific interfaces that enhance both data security and user productivity. They use advanced encryption techniques and access controls to protect data integrity and prevent unauthorized access.

JPMorgan also places a high emphasis on data security by implementing state-of-the-art cybersecurity measures. These include employing AI-driven security tools that monitor and detect anomalous activities in real-time, providing an additional layer of protection against cyber threats.

Actionable Advice

• Regular Audits: Conduct frequent audits of AI systems to ensure compliance and address potential risks.
• AI Integration: Integrate AI risk management into existing corporate frameworks for a seamless approach.
• Employee Training: Provide ongoing training on AI ethics and regulatory compliance to foster a culture of accountability.
• Data Protection: Implement advanced encryption and monitoring tools to safeguard data privacy and security.

By adopting these risk mitigation strategies, Goldman Sachs and JPMorgan not only protect their operations but also reinforce their positions as leaders in the responsible adoption of AI technologies.

Governance

In the rapidly evolving landscape of artificial intelligence, effective governance structures are crucial for ensuring responsible and ethical AI deployment. Both Goldman Sachs and JPMorgan have emerged as frontrunners in AI adoption, each demonstrating robust frameworks that enhance their business operations while adhering to ethical standards. This section delves into the governance practices of these financial giants, emphasizing their AI governance frameworks, leadership roles, and commitment to ethical AI practices.

AI Governance Frameworks

Goldman Sachs and JPMorgan have established comprehensive AI governance frameworks that provide a structured approach to AI deployment and management. These frameworks are designed to ensure compliance, transparency, and accountability. For instance, Goldman Sachs employs a centralized, multi-model AI platform that incorporates technologies from industry leaders like OpenAI and Google. This facilitates risk mitigation and ensures rapid adaptation and flexibility across business functions. According to recent reports, Goldman Sachs has successfully rolled out AI assistants to over 46,000 employees globally, showcasing their commitment to enterprise-wide AI deployment.

JPMorgan, on the other hand, emphasizes a decentralized AI governance model that allows individual departments to tailor AI solutions to their specific needs while adhering to overarching guidelines. This approach fosters innovation and agility, enabling JPMorgan to swiftly respond to emerging market trends and customer demands. A recent survey indicated that over 70% of JPMorgan's AI projects are initiated at the departmental level, demonstrating the effectiveness of their governance model in promoting innovation at the grassroots.

Role of Leadership in AI Strategy

Leadership plays a pivotal role in shaping AI strategies and ensuring their alignment with organizational goals. At Goldman Sachs, the leadership team actively champions AI initiatives by integrating AI into the company's core strategy. The CEO's commitment to AI as a competitive differentiator underscores the importance of leadership in AI adoption. Similarly, JPMorgan's board of directors has instituted an AI steering committee that oversees the strategic direction of AI projects, ensuring they align with corporate values and objectives.

Effective AI governance requires that leaders not only advocate for AI adoption but also invest in talent development and infrastructure. Both Goldman Sachs and JPMorgan have made significant investments in AI talent, with Goldman Sachs hosting quarterly workshops to upskill employees and JPMorgan allocating a substantial portion of its R&D budget to AI initiatives. Statistically, companies with strong leadership commitment to AI see a 20% increase in AI project success rates.

Ensuring Ethical AI Practices

Ensuring ethical AI practices is a cornerstone of AI governance for both firms. Goldman Sachs has implemented a set of ethical guidelines that govern AI development and deployment, focusing on fairness, accountability, and transparency. These guidelines are regularly reviewed and updated to reflect evolving ethical standards and legal requirements. Similarly, JPMorgan has committed to an ethical AI framework that prioritizes data privacy and bias mitigation. The bank employs advanced data anonymization techniques and conducts regular bias audits to ensure their AI systems operate impartially.

Actionable advice for organizations looking to enhance their AI governance includes establishing clear ethical guidelines, fostering a culture of continuous learning and adaptation, and involving diverse stakeholders in AI strategy development. By doing so, organizations can not only enhance their AI capabilities but also build trust with stakeholders and the broader public.

In conclusion, Goldman Sachs and JPMorgan exemplify best practices in AI governance, balancing innovation with ethical responsibility. As the AI landscape continues to evolve, their governance frameworks provide a valuable blueprint for other organizations aiming to harness the transformative potential of AI.

Metrics and KPIs: Evaluating AI Success and Transformation at Goldman Sachs and JPMorgan

As industry leaders in AI adoption by 2025, Goldman Sachs and JPMorgan have set benchmarks for assessing the success and impact of AI initiatives. Key Performance Indicators (KPIs) are crucial in measuring and benchmarking AI performance, as well as tracking AI-driven transformation within these financial giants. This section explores the metrics that define AI effectiveness and offers actionable insights for organizations aiming to leverage AI for competitive advantage.

Key Performance Indicators for AI Success

Goldman Sachs and JPMorgan employ a comprehensive set of KPIs to evaluate AI-driven initiatives. These indicators are designed to measure both qualitative and quantitative aspects of AI deployment:

• Operational Efficiency: AI's impact on reducing time and cost in operations. For example, Goldman Sachs has reported a 30% improvement in processing efficiency through AI automation.
• Revenue Growth: Contribution of AI to new revenue streams and enhanced customer experiences. JPMorgan's AI-driven analytics have increased cross-selling opportunities, boosting revenue by 15% in targeted segments.
• Risk Management: The effectiveness of AI in identifying and mitigating risks, such as fraudulent transactions or market volatility. Both firms have integrated AI to enhance risk prediction accuracy by 25%.
• Customer Satisfaction: Improvement in customer engagement and satisfaction scores, with AI providing personalized banking experiences. Goldman Sachs reported a 20% increase in Net Promoter Score (NPS) following AI assistant deployment.

Benchmarking AI Performance

Benchmarking against industry leaders like Goldman Sachs and JPMorgan involves assessing AI maturity and capabilities across several dimensions:

• Scalability: The ability to scale AI applications across the organization. Goldman Sachs has achieved enterprise-wide AI deployment, a benchmark for scalability.
• Adaptability: Flexibility in AI models to adapt to different business functions. JPMorgan's use of multi-model AI platforms ensures adaptability and responsiveness to market changes.
• Security and Compliance: Ensuring AI systems meet regulatory standards and protect data privacy, crucial for maintaining trust in financial services.

Tracking AI-Driven Transformation

Tracking the transformation brought by AI is essential for continuous improvement and strategic alignment. Here are some actionable strategies:

• Continuous Monitoring: Implement real-time dashboards for tracking AI performance metrics, allowing for proactive adjustments and optimizations.
• Employee Feedback: Collect feedback from employees to tailor AI tools to specific roles and enhance user experience. Goldman Sachs customizes interfaces based on role-specific needs, leading to higher productivity.
• Regular Audits: Conduct periodic audits to assess AI's alignment with business objectives and its impact on organizational goals.

In conclusion, the rigorous application of metrics and KPIs enables organizations like Goldman Sachs and JPMorgan to not only measure AI success but also drive continuous improvement. By adopting similar benchmarking practices, businesses can ensure their AI initiatives bring substantial strategic value.

Conclusion

As we conclude our comparison of Goldman Sachs and JPMorgan's AI adoption strategies, several key insights emerge. Both banks have demonstrated exemplary leadership in AI integration, underpinned by significant investment in technology, talent, and infrastructure. Goldman Sachs's approach, characterized by rapid enterprise-wide AI deployment, contrasts slightly with JPMorgan’s more gradual integration, allowing for lessons learned and iterative improvements.

Goldman Sachs’s strategy of deploying AI assistants to its entire 46,000+ global workforce highlights its commitment to leveraging AI beyond mere automation. Their centralized, multi-model AI platform not only ensures risk mitigation but also allows for rapid adaptation and flexibility. This strategic move has facilitated improved productivity, particularly by tailoring AI tools to specific job roles and workflows, yielding measurable business outcomes across various functions.

Conversely, JPMorgan's methodical roll-out, while slower, emphasizes sustainable growth and a robust framework to support long-term AI advancements. By focusing on data integrity and ethical AI practices, JPMorgan positions itself to harness AI responsibly, ensuring solutions not only meet current needs but are also scalable for future innovation.

Looking ahead, the future of AI in banking is set to expand, with trends pointing towards more personalized customer experiences, enhanced cybersecurity measures, and increased regulatory compliance through AI-driven analytics. Both Goldman Sachs and JPMorgan are poised to lead these advancements, setting benchmarks for AI adoption within the financial sector.

In conclusion, the AI journey of these two banking giants underscores the importance of strategic vision, adaptable technology, and a skilled workforce. As they continue to navigate the complexities of AI, other financial institutions can draw lessons from their experiences—particularly in prioritizing comprehensive risk management and fostering an innovation-centric culture. Institutions aiming to replicate their success should focus on aligning AI strategies with broader organizational goals, ensuring technological solutions are both effective and ethical.

Frequently Asked Questions