Category: AI

Using Copilot’s ‘Think Deeper’ option

Posted in AITagged , , , , ,

Microsoft Copilot Chat has introduced the ‘Think Deeper’ option (which has been available for free since March 2025). This is a huge to have OpenAI’s o1 model without needing a premium plan.

Using a recent news article, we used these two prompts consecutively:

summarise this page and analyse what it could mean for the digital and data market in new zealand based on the customer base and digital assets of trademe and stuff digital.

create a financial model outlining the financial breakeven point for this acquisition to provide a return.

Given the limited public information, the below results are based on significant assumptions but can be refined with additional information. Hopefully this shows how it could be applied to an internal problem with far more information available.

Below are the outputs from Copilot

Note: Not to be taken as facts or research but certainly shows the ease of use of the ‘Think Deeper’ model.

Summary of the TradeMe and Stuff Digital Acquisition

The article reports on a major strategic move in New Zealand’s digital landscape: Trade Me is taking a 50% stake in Stuff Digital, the digital arm of Stuff Group. Under the deal, Trade Me will gain a significant equity interest in the operation that runs high-traffic websites such as stuff.co.nz and ThreeNews. Notably, the agreement excludes Stuff Group’s traditional masthead publishing, events business, and certain other assets like Neighbourly. A key operational change is that the property section on Stuff Digital’s platform will be rebranded as Trade Me Property, with listings, advertisements, and selected content being shared between the two platforms. Both companies emphasize that while the partnership paves the way for enhanced growth opportunities and technological investments, editorial independence and ethical standards will remain paramount—with Stuff Digital retaining operational control via a board structure that ensures shared representation and a casting vote mechanism. This deal also marks an important milestone for Stuff, whose owner Sinead Boucher, having acquired the business for a nominal fee years ago, now sees a multi-million-dollar payday while bolstering future investments in technology and talent .

Analysis: Implications for the Digital and Data Market in New Zealand

1. Convergence of Digital Ecosystems and Enhanced Data Capabilities
This partnership represents a significant example of digital convergence in New Zealand. Trade Me, known for its expansive and highly engaged marketplace covering categories like property, motors, and jobs, brings an impressive customer base and deep transactional data. On the other hand, Stuff Digital dominates digital content and news consumption, offering extensive engagement metrics and rich consumer behavior insights. Combining these disparate yet complementary data sets could lead to powerful analytics capabilities, enabling more personalized content delivery and targeted advertising. Agencies and advertisers may gain unprecedented insight into consumer patterns, leading to innovations in tailored marketing strategies and data-driven service offerings .

2. Integrated Customer Experience and Cross-Market Synergies
For the digital market, the integration means that New Zealand consumers can expect a more seamless online experience. For example, the new Trade Me Property branding on Stuff Digital’s platform will likely enhance the discovery process for property buyers and sellers by leveraging Trade Me’s trusted marketplace interface. Such cross-pollination is expected to not only increase engagement on both platforms but also deliver more holistic digital experiences—from news to transactions—thereby reinforcing customer loyalty across New Zealand’s leading digital services .

3. Competitive Dynamics and Market Innovation
The deal signals an intensifying competitive landscape in New Zealand’s digital and data market. As two well-established Kiwi brands merge their digital assets, rivals in both the classifieds and media sectors may feel compelled to rethink their strategies. The fusion of expansive user data and a broad digital reach sets a new precedent for innovation, where data intelligence drives everything from content curation to marketplace recommendations. However, this also means that competitors will need to invest heavily in technology and consumer insights to keep pace, potentially spurring an overall boost in sector-wide innovation .

4. Data Privacy, Security, and Regulatory Considerations
With the merging of diverse datasets—from transaction records on Trade Me to content consumption metrics on Stuff Digital—there is enormous potential for deriving detailed consumer insights. However, this also elevates the stakes in terms of data privacy and security. Greater data integration can attract regulatory attention, meaning both companies must continue to prioritize user data protection, transparent policies, and robust ethical safeguards. Ensuring consumer trust in an era of heightened data sensitivity will be critical for sustaining long-term growth and market confidence .

5. Broader Digital Transformation in New Zealand
Finally, this partnership is emblematic of New Zealand’s broader digital transformation. As large-scale digital assets are consolidated and leveraged for more effective consumer engagement and data analytics, the move could set the stage for further investments in tech innovation across the country. The combined force of a popular marketplace and a trusted digital news source may well become a model for future collaborations, not only within New Zealand but also as an example for similar mid-sized markets globally .

Overall, the move by Trade Me to acquire a 50% stake in Stuff Digital heralds a new chapter in New Zealand’s digital ecosystem. It underscores the growing importance of integrating commerce and content, the value of robust data assets, and the need for strategic innovation in delivering consumer experiences. As this merger evolves, one can expect ripple effects across advertising, customer engagement, and data analytics—transforming the way digital businesses operate in New Zealand and beyond.

Below is a conceptual financial model that outlines how Trade Me’s 50% acquisition in Stuff Digital might reach a breakeven on its investment. Keep in mind that actual figures would need to be sourced from internal company data and market research; the numbers below are hypothetical yet illustrative of the underlying dynamics.

Financial Model – Key Assumptions

  1. Acquisition Cost: For this model, we assume Trade Me pays NZD 50 million for a 50% stake. (This implies an overall enterprise value for Stuff Digital of NZD 100 million, though our focus here is on Trade Me’s outlay.)
  2. Initial Revenue & Margins:
    • Year 0 (pre-integration) annual revenue from the Stuff Digital platform is assumed at NZD 25 million.
    • The standalone operating margin is modest at 10%, resulting in an operating profit of NZD 2.5 million.
  3. Growth and Synergy Effects:
    • Revenue Growth: Integration synergies (e.g., cross-promotional opportunities between Trade Me’s marketplace and Stuff Digital’s content channels) spur an enhanced annual revenue growth rate of about 10% per year (a blend of organic growth plus synergy lift).
    • Margin Improvement: With cost efficiencies in operations and improved digital capabilities, operating margins are expected to rise to approximately 15% from Year 1 onward.
  4. Timeline: We will track annual operating profit accumulation over a 10-year period until cumulative profit equals the NZD 50 million investment. (Note that discounting for the time value of money is not shown here but can be incorporated through a discounted cash flow approach once more detailed projections are available.)

Simplified Financial Model

The following table illustrates a year-by-year projection under the above assumptions:

YearRevenue (NZD million)Operating MarginAnnual Operating Profit (NZD million)Cumulative Operating Profit (NZD million)
025.010%2.502.50
125.0 × 1.10 = 27.515%27.5 × 0.15 = 4.132.50 + 4.13 = 6.63
227.5 × 1.10 ≈ 30.2515%30.25 × 0.15 ≈ 4.546.63 + 4.54 ≈ 11.17
330.25 × 1.10 ≈ 33.2815%33.28 × 0.15 ≈ 5.0011.17 + 5.00 ≈ 16.17
433.28 × 1.10 ≈ 36.6115%36.61 × 0.15 ≈ 5.4916.17 + 5.49 ≈ 21.66
536.61 × 1.10 ≈ 40.2715%40.27 × 0.15 ≈ 6.0421.66 + 6.04 ≈ 27.70
640.27 × 1.10 ≈ 44.3015%44.30 × 0.15 ≈ 6.6427.70 + 6.64 ≈ 34.34
744.30 × 1.10 ≈ 48.7315%48.73 × 0.15 ≈ 7.3134.34 + 7.31 ≈ 41.65
848.73 × 1.10 ≈ 53.6015%53.60 × 0.15 ≈ 8.0441.65 + 8.04 ≈ 49.69
953.60 × 1.10 ≈ 58.9615%58.96 × 0.15 ≈ 8.8449.69 + 8.84 ≈ 58.53
1058.96 × 1.10 ≈ 64.8515%64.85 × 0.15 ≈ 9.7358.53 + 9.73 ≈ 68.26

Interpretation: Up to Year 0 (pre-integration), the platform’s performance contributes a modest NZD 2.5 million in operating profit. From Year 1, as the enhanced model kicks in (with synergies and margin improvements), the cumulative operating profit grows steadily.

According to the table, the cumulative operating profit crosses the NZD 50 million threshold between Year 8 and Year 9. This suggests that, under these assumptions, the breakeven point (where the cumulative operating profit equals the acquisition cost of NZD 50 million) is reached in roughly 8–9 years.

Note that incorporating the time value of money via discounting could extend the break-even timeline, and additional factors (like integration costs, tax implications, capital expenditures, and unforeseen market shifts) may also impact the actual period required.

What This Means in the Broader Context

  • Revenue and Data Synergies: With Trade Me’s extensive customer base and robust marketplace data combined with Stuff Digital’s rich digital content platform, the model assumes that the companies can drive an accelerated revenue growth rate. This integration deepens customer engagement and opens up new advertising and cross-selling opportunities.
  • Cost Efficiencies: Margin improvements reflect both cost-saving synergies (e.g., shared technology and infrastructure investments) and better monetization of digital assets through targeted advertising and premium content offerings.
  • Market Leadership: Achieving the breakeven point within 8–9 years would not only validate the strategic investment but also create a financially stronger, more competitive presence in New Zealand’s digital landscape. This would likely spur further innovation and could set a benchmark for similar consolidations in mid-sized markets.
  • Risk Considerations: Realizing these synergies depends on successful integration, continual innovation, and maintaining strong user trust and data privacy standards. The assumptions in growth and margin improvement are key levers in the model; deviations here could shift the break-even timeline.

Divergent Insights

While this simplified model provides a framework, further analysis could delve into:

  • Sensitivity Analysis: Quantify how changes in revenue growth, margin improvement, or capital costs affect the break-even timeline.
  • Discounted Cash Flow (DCF): Apply an appropriate discount rate (e.g., 8–10% given market conditions) to assess the present value of future operating profits.
  • Scenario Modeling: Consider high‐growth vs. conservative scenarios to understand upside potential or downside risks in the digital and data market.
  • Integration Costs & Capital Expenditures: Factor in any additional investments needed for technology, talent, or integration activities which could delay breakeven.

Video creation tools

Posted in AI, AI AdoptionTagged , , , ,

With Google extending their Veo option, it’s always good to revisit the key options that are available.

Here’s a high-level look at each AI-powered video and audio generation tool from major tech companies, including links to explore them:

1. OpenAI – Sora

Link: Explore Sora

  • Cost: Available through ChatGPT Plus and Pro subscriptions, starting at around US$20 per month.
  • Quality: Generates high-quality, realistic videos up to 60 seconds long with complex scenes and camera movements.
  • Safety: OpenAI has implemented safeguards to prevent misuse, including red teaming and watermarking AI-generated content.

2. Adobe – Firefly

Link: Try Adobe Firefly

  • Cost: Subscription-based model integrated into Adobe Creative Cloud.
  • Quality: Offers professional-grade AI-generated videos, animations, and visual effects with seamless integration into Adobe Premiere Pro and After Effects.
  • Safety: Firefly is trained on Adobe Stock and public domain content, ensuring commercially safe outputs.

3. Google – Veo 3

Link: Use Veo 3

  • Cost: Available through Google AI Ultra subscription ($249.99/month) and Google AI Pro for limited access (8 second video).
  • Quality: Generates high-resolution videos with integrated audio, including dialogue and ambient sounds.
  • Safety: Google emphasizes AI ethics, prompt adherence, and safeguards against misuse.

4. Meta – MovieGen

Link: Learn about MovieGen

  • Cost: Currently in research phase, not yet commercially available.
  • Quality: Capable of generating 1080p HD videos with synchronized audio, personalized video creation, and precise editing.
  • Safety: Meta is developing safeguards to prevent misuse and ensure ethical AI deployment.

5. RunwayML

Link: Try RunwayML

  • Cost: Freemium model with paid tiers for advanced features.
  • Quality: Offers high-resolution AI-generated videos with advanced editing tools, including lip-syncing and generative audio.
  • Safety: Implements AI ethics policies to prevent harmful content.

Each platform has unique strengths, whether it’s affordability, professional quality, or ethical safeguards. The evolution in this space is rapid but does not yet replace the need for high quality media content.

Pitfalls

Here’s an example of a generated video that initially appears what I asked for. After a few seconds, the man on the left appears to sneeze and disappears (quite the magic trick!).

Prompt: Create a hot steaming cup of coffee with a colour changing data logo on the outside with a blurred background of an office with robots and humans sitting at desk wearing headsets.

Model: Google Veo 2

Agentic AI to drive efficiency, accuracy and communication

Posted in AI, Automation, Uncategorized

A new paradigm is emerging that promises to evolve how businesses operate: Agentic AI. This advanced form of AI is designed to autonomously pursue complex goals and workflows with minimal human intervention, driving efficiency, reducing AI hallucinations, and enabling AI-to-AI communication. Let’s explore how Agentic AI is set to transform the business world.

Understanding Agentic AI

Agentic AI refers to AI systems that exhibit autonomous decision-making, planning, and adaptive execution to complete multi-step processes. Unlike traditional AI, which is often limited to specific, predefined tasks, Agentic AI can understand and interpret complex contexts and goals, allowing for more sophisticated interactions and decision-making processes.

Key capabilities of Agentic AI include:-

  • Autonomy: The ability to take goal-directed actions with minimal human oversight.
  • Reasoning: Contextual decision-making to make judgment calls and weigh tradeoffs.
  • Adaptable Planning: Dynamic adjustment of goals and plans based on changing conditions.
  • Language Understanding: Comprehending and following natural language instructions.-
  • Workflow Optimisation: Fluidly moving between subtasks and applications to complete processes efficiently without silo boundaries.

Driving Business Efficiency

One major advantage of Agentic AI is its ability to enhance business efficiency. By managing complex workflows automatically, Agentic AI can optimise operations, cut costs, and adapt to changing business needs. For instance, in supply chain management, it can monitor inventory, predict demand, and adjust orders in real-time, minimising tock for manufacturers or out of stocks for retailers. In customer service, it handles queries, offers personalized responses, and escalates issues to human agents only when necessary, improving response times and customer satisfaction.

Reducing AI errors through controls

AI hallucinations, where AI systems generate incorrect or nonsensical information, have been a significant challenge in deploying AI solutions. Agentic AI addresses this issue by incorporating advanced reasoning and contextual understanding capabilities. These systems can cross-verify information, learn from past interactions, and adapt their responses based on real-world feedback, significantly reducing the likelihood of hallucinations.Moreover, implementing a human-in-the-loop system, where human agents verify AI responses before they are finalised, can further mitigate the impact of hallucinations. This approach not only improves the accuracy of AI outputs but also enhances the system’s learning over time.

Enabling AI-to-AI Communication

Another groundbreaking aspect of Agentic AI is its ability to facilitate AI-to-AI communication. In a multi-agent system, different AI agents can collaborate, share information, and coordinate actions to achieve common goals. This capability is particularly beneficial in complex environments where multiple systems (or even organisations) need to work together. For instance, in a smart manufacturing setup, AI agents controlling different parts of the production line can communicate to optimise the entire process. One agent might monitor the quality of raw materials, while another adjusts the machinery settings based on real-time data, ensuring that the final product meets the desired standards. This level of coordination can lead to significant improvements in efficiency, quality, and overall productivity.

The Future of Agentic AI

As Agentic AI continues to evolve, its applications will expand across various industries, driving innovation and efficiency. Businesses that adopt Agentic AI will be better positioned to navigate the complexities of the modern market, respond to emerging challenges, and capitalize on new opportunities. In conclusion, Agentic AI represents an important shift in how artificial intelligence can autonomously drive enterprise solutions. By enhancing business efficiency, reducing AI hallucinations, and enabling AI-to-AI communication, Agentic AI is poised to become an indispensable tool for organisations aiming to thrive in the digital age.

Understanding AI and ML in Modern Enterprise Solutions

Posted in AI, UncategorizedTagged , , , ,

In today’s rapidly evolving technology landscape, the integration of Artificial Intelligence (AI) and Machine Learning (ML) into enterprise applications has sparked both excitement and skepticism. While some argue that trusting AI/ML to handle critical tasks indicates a lack of concern for the job’s quality, others see it as a revolutionary step towards efficiency and innovation. Let’s delve into this debate and explore the nuanced roles of AI and ML in enterprise applications.

The Purpose of Enterprise Applications

Enterprise applications are designed to codify workflows and processes, ensuring consistency and reliability across various operations. These applications rely on well-defined algorithms to perform tasks that require precision and predictability. For instance, determining whether it’s freezing outside can be easily handled by mathematical formulas and sensors.However, there are scenarios where traditional algorithms fall short. This is where ML comes into play. ML is specifically designed to tackle problems that are difficult or impossible to solve with conventional algorithms. Examples include identifying spam emails, detecting nudity in images, or recognising pedestrians in real-time video feeds. These tasks require a level of adaptability and learning that static algorithms cannot provide.

The Misconceptions about AI and ML

A common misconception is that AI and ML are meant to replace well-codified processes. This is not the case. AI, particularly in its current state, is intended to augment existing algorithms, handling edge cases and providing insights where traditional methods fail. For example, while conventional code is deterministic and reliable, ML models are probabilistic and can adapt to new data patterns, making them suitable for dynamic and complex environments.Critics often label AI initiatives, such as those by OpenAI, as overhyped. They argue that these technologies are not designed to codify processes but to enhance them. It’s crucial to understand that AI and ML are tools that complement traditional coding, not substitutes for it. Conventional code remains superior for tasks that require deterministic outcomes, while ML excels in areas where adaptability and learning from data are essential.

The Challenges of Implementing AI and ML

Implementing AI and ML in enterprise applications is not without challenges. One significant hurdle is ensuring that the AI-generated solutions work reliably. Writing code that functions correctly, scales efficiently, and remains maintainable over time is a complex task. It requires rigorous testing and validation, especially when financial stakes are high.Moreover, maintaining AI-driven systems poses its own set of challenges. Business processes are rarely static; they evolve over time, necessitating continuous updates and modifications. While generating initial code might be straightforward, making precise changes without introducing errors is a daunting task. This is where the expertise of developers, analysts, and quality assurance professionals becomes indispensable.

Compute costs, Environmental and Economic Impact

Another critical aspect to consider is the cost, environmental and economic impact of AI and ML. Training and deploying ML models require substantial computational resources, leading to significant energy consumption and ecological concerns. Additionally, the costs associated with developing and maintaining AI-driven systems can be high, making it essential to weigh up the cost benefit.

AI’s Role in Creating ML Models

Interestingly, AI itself can be leveraged to create ML models. Automated Machine Learning (AutoML) platforms enable users to build ML models without extensive coding knowledge. These platforms use AI to automate the process of selecting algorithms, tuning hyperparameters, and validating models. This democratizes access to ML, allowing businesses to harness its power without requiring deep technical expertise.

Wrap up

AI and ML are powerful tools that, when used appropriately, can significantly enhance enterprise applications. They are not meant to replace traditional coding but to augment it, handling tasks that require adaptability and learning from data. While there are challenges and misconceptions surrounding their implementation, the potential benefits make them invaluable assets in the modern technological landscape. As we continue to innovate, the synergy between AI, ML, and conventional coding will drive the next wave of enterprise solutions.—What are your thoughts on the role of AI and ML in enterprise applications? Do you see them as complementary tools or potential replacements for traditional methods?

Innovation unlocked

Posted in AI, Automation, Modern Workplace

Convergence of MACH, Digital Dexterity, and AI unlocking unbounded innovation at pace

At the CIO Summit, we discussed convergence of three trends across integration, capability and intelligence so below is quick reflections on the topics.

MACH (Microservices, API-first, Cloud-native SaaS, Headless), Digital Dexterity, and AI. These elements are intertwine to create further human-digital integration and automation building on each other. Below we’ll cover them individually and then their combined strength to drive innovation.

MACH: The Backbone of our Digital World

MACH as a toolkit empowers us to build flexible, scalable, and efficient digital solutions. Microservices are like Lego bricks, allowing us to create modular applications with APIs as the glue communicating between different components. Cloud-native SaaS leverages the power of the cloud, providing scalability and flexibility. Headless architecture decouples the front-end from the back-end, offering greater customisation and agility particularly with AI.

MACH Alliance ensure technology is future-proofed and drive digital experiences

https://machalliance.org/the-mach-manifesto

Digital Dexterity: The Human Edge

While technology provides the foundation, it’s our team members’ digital dexterity that drives innovation. It’s about understanding how technology can solve problems, leading our organisations through the digital transformation journey, and continuously exploring new ways to leverage business value.

https://hbr.org/2021/10/how-to-build-digital-dexterity-into-your-workforce

https://quixy.com/blog/basics-of-digital-dexterity/

AI: Our (mostly) Intelligent Assistant

AI is rapidly evolving, transforming industries and reshaping the way we work. From chatbots to algorithms that can diagnose diseases, AI is becoming increasingly sophisticated. It’s augmenting our capabilities, automating routine tasks, providing insights, and making business truly global.

Convergence: A Powerful Force for Innovation

When we combine MACH, Digital Dexterity, and AI, we create a powerful force for innovation. This convergence is leading to:

  1. Hyper-Personalised, Evolving Experiences: We can deliver highly tailored experiences to our customers, that continually evolve.
  2. Intelligent Self-Healing Automation: Routine tasks can be automated across organisational boundaries, freeing up time to focus on more strategic and creative activities.
  3. Continuous Innovation: The modular nature of MACH architecture and the power of our people, supported by AI, enable us to adapt quickly (potentially automatically).

Exploring the possibilities

The future is full of potential. What if AI-powered virtual assistants that could anticipate our needs? Or imagine developing intelligent applications that could optimise themselves to reduce costs or find more data? Or brainstorming article topics, peer reviewing and publishing in different languages and formats based on customer signals and searches?

Questions for us to consider

As we navigate this potential, it’s important to ask ourselves:

✅️ How can we ensure that AI is used ethically and responsibly?

💖 How can we prepare our teams for the skills of the future?

🔥 What are the potential risks and challenges relying heavily on technology to automate?

The convergence of MACH, Digital Dexterity, and AI is shaping a new era of innovation. By embracing these trends, we can create a bright future (or a scary one if we get it wrong).

AI Use Cases

Posted in AI, AI Adoption

Below is a couple of useful areas:

  1. Approach to identify & prioritise AI use case
  2. Suggested AI use cases
  3. Potential growth of AI and improvements over the next 3 years

Approach to Identify and Prioritise AI Use Cases

Identifying and prioritising AI use cases is crucial for maximising business value and ensuring successful AI implementation. Here’s a structured approach inspired by Gartner’s methodologies (Gartner Toolkit).

1. Ideation and Identification

  • Crowdsource Ideas: Gather input from various stakeholders, including employees, customers, and partners. Use workshops, brainstorming sessions, and surveys to collect a wide range of potential AI use cases.
  • Strategic Alignment: Ensure that the identified use cases align with the organization’s strategic goals and objectives. Focus on areas that can drive significant business value and support long-term strategies.
  • Pain Point Analysis: Identify critical business problems and pain points that AI can address. Prioritize use cases that solve high-impact issues and offer substantial improvements in efficiency or customer satisfaction.

2. Assessment and Feasibility

  • Technical Feasibility: Evaluate the technical requirements and feasibility of each use case. Consider data availability, quality, and the complexity of AI models needed.
  • Business Impact: Assess the potential business impact of each use case. Estimate the expected ROI, cost savings, revenue generation, and other tangible benefits.
  • Resource Availability: Determine the resources required, including budget, talent, and technology. Ensure that the organization has or can acquire the necessary resources to implement the use case (or choose a different first use case).

3. Prioritisation Framework

  • Scoring System: Develop a scoring system to rank use cases based on criteria such as strategic alignment, business impact, technical feasibility, and resource availability. Assign weights to each criterion based on its importance to the organisation.
  • Quick Wins vs. Big Bets: Categorize use cases into “quick wins” (low effort, high impact) and “big bets” (high effort, high impact). Prioritize quick wins to demonstrate early success and build momentum but also outline big bets to build on early success.
  • Risk-Reward Analysis: Conduct a risk-reward analysis to evaluate the potential risks and rewards associated with each use case. Focus initially on the low risk options to minimise hurdles.

4. Implementation and Monitoring

  • Pilot Projects: Start with pilot projects to test the feasibility and impact of prioritised use cases. Use the results to refine the approach and scale successful pilots (and take lessons from unsuccessful pilots to enhance future success).
  • Continuous Monitoring: Implement a continuous monitoring system to track the performance and impact of AI use cases. Use metrics and KPIs to measure success and identify areas for improvement through an objective lense.
  • Iterative Improvement: Regularly reassess and reprioritize AI use cases based on new data, changing business needs, and technological advancements. Maintain flexibility to adapt to evolving circumstances.

By following this structured approach, organizations can effectively identify and prioritise AI use cases that align with their strategic goals, maximize business value, and ensure successful AI implementation.

If you’re stuck for ideas, below is an outline of a few ideas to get you started.

Potential Business Use Cases for AI

Artificial Intelligence (AI) is revolutionising various industries by enhancing efficiency, reducing costs, and driving innovation. Here, we explore some of the key business use cases for AI and the potential for growth in the next three years.

1. Customer Service and Support

AI-powered chatbots and virtual assistants are transforming customer service by providing 24/7 support, handling routine inquiries, and improving response times. Companies like Amazon and Netflix use AI to personalize customer interactions, leading to increased satisfaction and loyalty.

2. Cybersecurity and Fraud Detection

AI is crucial in identifying and mitigating cyber threats. Machine learning algorithms analyse patterns to detect anomalies and potential security breaches. Businesses are increasingly adopting AI for cybersecurity, with over half of businesses using AI in these areas.

3. Personalised Marketing

AI enables businesses to tailor marketing campaigns to individual customer preferences. By analysing data from various touchpoints, AI can predict customer behaviour and recommend products. This approach has significantly boosted sales for companies like Amazon and Sephora.

4. Supply Chain Optimization

AI helps in optimising supply chain operations by predicting demand, managing inventory, and identifying inefficiencies. This leads to cost savings and improved operational efficiency. AI-driven supply chain management is expected to grow significantly as businesses seek to enhance their logistics.

5. Human Resources and Recruitment

AI streamlines the recruitment process by automating resume screening, scheduling interviews, and even assessing candidate fit through predictive analytics. This not only saves time but also improves the quality of hires. AI in HR is projected to grow as companies look to enhance their talent acquisition strategies.

6. Healthcare and Diagnostics

AI is making significant strides in healthcare by assisting in diagnostics, personalized treatment plans, and patient monitoring. AI algorithms can analyse medical images, predict disease outbreaks, and even suggest treatment options. The healthcare AI market is expected to see substantial growth as technology continues to advance.

7. Financial Services

In the financial sector, AI is used for fraud detection, risk management, and personalized banking services. AI-driven algorithms can analyse transaction patterns to detect fraudulent activities and assess credit risks. The adoption of AI in finance is anticipated to increase as institutions seek to enhance security and customer experience.

Potential for AI Growth in the Next 3 Years

The AI market is poised for significant growth in the coming years. According to various industry reports, the global AI market is expected to grow at a compound annual growth rate (CAGR) of over 35% from 2023 to 2026. This growth is driven by advancements in AI technology, increased adoption across industries, and the continuous need for automation and efficiency.

Key Drivers of Growth:

  1. Technological Advancements: Continuous improvements in AI algorithms, computing power, and data availability are driving the adoption of AI.
  2. Increased Investment: Businesses are investing heavily in AI to gain a competitive edge, leading to rapid innovation and deployment.
  3. Regulatory Support: Governments worldwide are recognising the potential of AI and are implementing policies to support its growth.
  4. Consumer Demand: As consumers become more accustomed to AI-driven services, their demand for personalized and efficient solutions is increasing.

In conclusion, AI is set to play a pivotal role in transforming business operations across various sectors. The next three years will likely see accelerated adoption and innovation, making AI a key tool for businesses aiming to remain competitive.

April 2024 Key Learnings

Posted in AI, Data, Modern Workplace

There is so much going on in the world of Data + AI, I thought I would summarise and challenge myself to a new format! This is my personal perspective informed by global research and experience.

⚠ Emerging Risks can no longer be ignored, impacting daily tasks
🤖 AI is nearing or exceeding human performance for many tasks
😟 AI Anxiety is increasing but not related to likely impacts to roles
👋 Business users need AI to power actions across the organisation
💾 All layers of data and techniques need to be considered
📈 Optimising LLMs can be expensive but key gains achieved easily

It’s not quite the end of the month but nothing interesting ever happens on a Tuesday 😁.

dataai.nz-key-learnings-april-2024Download

AI Policy, Principles & Usage Standards

Posted in AI, AI Governance

Key steps

When you’re ready to produce your company AI guidelines, here are some steps you can follow to plan, create, and implement your protocols.

  1. Key steps
    1. Establish a working group of broad stakeholders
    2. Define AI, Purpose & Scope
    3. Establish your core AI principles
    4. Assess AI Risks and Compliance needs
    5. Set accountability
    6. Design monitoring, control and review procedures
    7. Plan & Implement
    8. Develop best practice standards & training
  2. What is out there today?
    1. Google AI Principles
      1. What is allowed?
      2. What is not allowed?
    2. OECD AI Principles
      1. Values-based principles
      2. Recommendations
    3. Spark NZ AI Principles
  3. Wrap Up

There are some great guidelines out there for this process. Below are the two extremes from corporate governance to startups but follow a similar pattern. Below is a summary of a path that you may be able to follow.

Establish a working group of broad stakeholders

Create a group of people to lead the development of your AI policy, principles and usage standards. This working group should include leadership and may even include board members, executives, and department heads, as well as relevant technical experts. They should have a foundational understanding of AI and its benefits and risks (or may require this education as the first step).  You need to gather diverse insights and ensure the policy addresses all relevant aspects.

Define AI, Purpose & Scope

Create a common understanding of what is AI. It’s important that you create a definition of AI that suits your organisation as this is the foundation for an AI policy, principles or standard so your teams are clear on what is considered AI and what is not.

Identify the ways in which you will use AI in your operations or develop AI technologies. Define the overarching goals of AI use within your organisation and the boundaries of its application (including what you won’t do).

Within the scope, you should be clear on the components of what is required – for example, policies, principles, standards, and/or procedures. This will also help determine who is required to help.

Establish your core AI principles

Determine how your company’s core principles will relate to AI use and development. These principles will inform the ethical foundation of your AI usage policy. Some initial focus areas should be on autonomy, bias, security, and privacy.

Assess AI Risks and Compliance needs

Learn about the legislation that govern AI tools and ensure that your guidelines are compliant with these regulations. Work to draft legislation or guidance to design with future compliance in mind as much as possible.

Identify potential risks associated with AI usage and ensure compliance with existing policies, legal and ethical standards. Ethical principles that will guide the development and deployment of AI, such as fairness, transparency, and privacy.

There are some great frameworks out there and our preferred starting point is the EU AI Act categorises AI systems into four risk levels:

  1. Unacceptable Risk: Applications that considered a threat to people.
    These comprise subliminal techniques, exploitative systems or social scoring systems are banned. Also prohibited are any real-time remote biometric identification systems (e.g. such as facial recognition) with some exceptions for law enforcement purposes.

    These are banned, unless permitted in rare cases for law enforcement purposes or with a court order.
  2. High Risk: These are AI systems that are subject to significant regulatory obligations – primarily products that meet safety laws and software used to manage human facilities or health.

    This requires enhanced thresholds of diligence, initial risk assessment, and transparency. The technology itself will need to comply with certain requirements – including around risk management, data quality, transparency, human oversight, and accuracy.
  3. Limited Risk: These are AI systems will have to comply with transparency requirements, copyright law, showing the content was generated by AI and designs to prevent it from generating illegal content.
  4. Minimal or No Risk: These AI systems likely have the least regulatory oversight.

There is also a potential category for Systemic Risk for high-impact general-purpose AI models like ChatGPT-4. This is inline with any system (banking, insurance or infrastructure).

Each risk level has corresponding regulatory requirements to ensure that the level of oversight is appropriate to the risk level. The Act adopts a risk-based approach, aiming to boost public confidence and trust in technology.

Set accountability

Create a structure for decision-making, accountability, and oversight of AI systems.

Determine who will be responsible for which stages of selection, use, development and monitoring of AI tools in your organisation. Create processes for reporting and governance so that your usage guidelines are followed as intended.

Design monitoring, control and review procedures

Establish procedures for monitoring AI performance and managing deviations from the policy. These procedures will enable you to monitor how your AI systems are performing and following your agreed policies. You should also regularly evaluate your policies to ensure they remain effective at mitigating risks and evolve with changes in processes and technology. 

Plan & Implement

When the guidelines are complete, they need to be communicated to all stakeholders clearly, so all understand them (with a potential test). Provide training for employees on responsible AI usage and the implications of the AI policy. Given the impacts across most workforces, this could incorporate the full company as AI users with a focus group for AI designers or implementors.

Once this is done you should have a process to monitor and update the policy.

Develop best practice standards & training

Create standards for using AI while mitigating risk and balancing human and artificial intelligence appropriately. Outline key scenarios, for example customer risks, to help teams understand what is required.

What is out there today?

Google AI Principles

Google have a very clear set of principles outlined on their AI site: https://ai.google/responsibility/principles/

Below is a summary based on the website as at 23 March 2024. The structure is very easy to understand and read with broad outcomes that they want to focus on and also outline where they will not participate (including acknowledging this has and will change as they learn more).

What is allowed?

  • Be socially beneficial.
  • Avoid creating or reinforcing unfair bias.
  • Be built and tested for safety.
  • Be accountable to people.
  • Incorporate privacy design principles.
  • Uphold high standards of scientific excellence.
  • Work to limit potentially
    harmful or abusive applications.

What is not allowed?

  • Cause or are likely to cause overall harm.
  • Weapons or other technologies whose principal purpose is to cause injury.
  • Surveillance violating internationally accepted norms.
  • Contravenes widely accepted principles of international law and human rights.
  • This list may evolve.

Google also publish their principles back to 2019, which have largely stayed consistent.

OECD AI Principles

Below is a summary based on the website (https://oecd.ai/en/ai-principles) as at 23 March 2024.

The OECD AI Principles promote use of AI that is innovative and trustworthy and that respects human rights and democratic values. The principles represent a common aspiration that can shape a human-centric approach to trustworthy AI.

Values-based principles

  • Inclusive growth, sustainable development and well-being
  • Human-centred values and fairness
  • Transparency and explainability
  • Robustness, security and safety
  • Accountability

Recommendations

  • Investing in AI R&D
  • Fostering a digital ecosystem for AI
  • Providing an enabling policy environment for AI
  • Building human capacity and preparing for labour market transition
  • International co-operation for trustworthy AI

As an organisation focussed on policy development, they are focussed more on the recommendations and actions that countries and policymakers should take. Their recommendations highlight areas that a company also needs to focus on around investment, delivering on a digital ecosystem, outlining policies and providing upskilling and transition for current team members.

Spark NZ AI Principles

Spark have committed and made their principles public which shows great leadership within NZ corporates. There policy is available here:

Spark NZ AI Principles (December 2023).pdf

Artificial Intelligence (AI) technologies are evolving and being deployed at scale. These
technologies have increasingly sophisticated capabilities, some of which can directly impact
people or influence their behaviours, opinions, and choices.

  • Human centred
  • Ethical design
  • Diversity, inclusivity, and bias
  • Safety and reliability
  • Privacy
  • Informed human decision making
  • Explicability and transparency

Interestingly unlike Google, Spark do not outline what they will not do, more how they will be in line with their ethics and requirements.

Wrap Up

There are many resources available as a starting point for getting this underway at your company – you just need to start. Make sure that you engage broad stakeholders to give you great inputs on the future needs of your business to deliver on responsible AI (or no AI if that’s the business decision and risk appetite). We look forward to seeing many more companies in New Zealand publish their approach and share with their customers and key stakeholders for a more transparent future for us all.

What is AI?

Posted in AI, AI Adoption

Overview

It’s important that you create a definition that suits your organisation as this is the foundation for an AI policy, principles or standard so your teams are clear on what is considered AI and what is not.

Dynamic

Within these discussions, it’s often to start with a definition of AI as this has evolved over the past 5 years considerably. The OECD shared their evolution here: https://oecd.ai/en/wonk/definition

It is important that your definition covers the current state but also considers the significant advancements, such as into all types of media, coming in the near future.

Communicating

It is important to have a clear way of communicating Artificial Intelligence (AI) across your organisation for your own team, leadership and Board. The uses of AI within your organisation may also need to be shared with stakeholders, shareholders and regulators.

EU Parliament have this for everyday use: https://www.europarl.europa.eu/topics/en/article/20200827STO85804/what-is-artificial-intelligence-and-how-is-it-used

Potential Definitions

The OCED (2023) definition of AI

An AI system is a machine-based system that, for explicit or implicit objectives, infers, from the input it receives, how to generate outputs such as predictions, content, recommendations, or decisions that can influence physical or virtual environments. Different AI systems vary in their levels of autonomy and adaptiveness after deployment.

This being from a policy organisation using complex words and can be quite hard to understand for most people (although also highly accurate).

The EU definition of artificial intelligence (AI)

AI is the ability of a machine to display human-like capabilities such as reasoning, learning, planning and creativity.

AI enables technical systems to perceive their environment, deal with what they perceive, solve problems and act to achieve a specific goal. The computer receives data – already prepared or gathered through its own sensors such as a camera – processes it and responds.

AI systems are capable of adapting their behaviour to a certain degree by analysing the effects of previous actions and working autonomously.

https://www.europarl.europa.eu/topics/en/article/20200827STO85804/what-is-artificial-intelligence-and-how-is-it-used

The EU definition is pretty good outlining the human-like capabilities but references to a “computer” is slightly unusual as it’s likely to be from a computer, ecosystem or the internet. Also it’s questionable whether all AI is acting to achieve a specific goal – that is more similar to machine learning.

The 10 year old definition of AI

An AI system is a smart machine.

  • It gets information (input) and uses that to do things (actions) like creating stuff (text, images, video), giving advice or making choices.
  • These actions can change things in the real or digital world.
  • It may work on their own (or require humans).
  • It can learn from their experiences after they are set up.
  • Each AI system can do this differently.

There is no single definition of AI and this will continue to evolve (no doubt even the above examples from March 2024 will look dated soon).

Wrap Up

It’s important that you create a definition that suits your organisation as this is the foundation for an AI policy, principles or standard so your teams are clear on what is considered AI and what is not.