Our experts discuss real AI wins for non-technical business owners who barely have time to breathe, this week on Let’s Talk.
Business owners are drowning in AI promises but can’t find 10 minutes to figure out where to start, let alone implement anything that actually works.
The key is to ignore the noise about complex enterprise AI and focus on practical tools that solve real problems right now. The biggest AI mistake business owners make is overthinking where to start. Forget enterprise chatbots and machine learning models begin with AI tools that require zero technical knowledge.
Here’s how our experts are discovering AI tools that actually work without technical expertise and deliver real results in this week’s Let’s Talk.
Let’s Talk!
Contribute to Dynamic Business ✍
Sonia Eland, Executive Vice President and Country Manager, Australia and New Zealand, HCLTech
“You don’t need to be a coder to use AI effectively – just be clear on what you want it to do. Start by identifying one or two routine tasks that drain your time, such as summarising documents, drafting standard emails, or organising information. Many tools you already use have AI built in, so you can start experimenting without adding new software.
“Even a small investment in AI literacy pays off quickly. Short, practical resources now make it easier for non-technical professionals to understand and apply AI. At HCLTech, we’ve seen that when people gain adaptable, future-ready skills, they not only understand AI but also trust themselves to use it. That confidence is what turns AI from a buzzword into a practical tool.
“The next step is to start treating AI like a colleague: give clear instructions, review its work, and refine the results. Begin with small wins and build from there. AI can take on repetitive work, freeing you to focus on strategy, creativity, and problem-solving.
“I’d encourage you to explore AI platforms to enhance your work but always be mindful of what you upload. These tools may sit outside your firewall, so use discretion when handling sensitive information.”
Asanga Lokusooriya, IBM Consulting Lead Client Partner, IBM Consulting
“If you’re not technical and short on time, don’t think of AI as a new skill to learn but think of it as an extension of how you already work. Start with the tools you use every day like Outlook, Teams, Excel, Word, etc., many now have AI built in. Use it to strip away low-value work: summarise documents, distil meetings into action points, draft responses, and prepare talking points.
Give AI a clear brief: context, what you need, and the format you want just as you would with a team member. Pick one or two AI capabilities that make the biggest difference to your output and repeat them until they’re second nature. Always apply your judgment as AI gives you speed and options, but you provide the insight and nuance. Protect sensitive information by sticking to enterprise-approved solutions. The goal is not to “do AI” but to integrate it into your flow so it quietly multiplies your capacity. In short, use AI to help you think, decide, and deliver faster without adding another layer of complexity to your day.”
Shaun Broughton, Managing Director, APAC and Japan, Shopify
“AI isn’t just for coders or big tech teams—it’s a practical tool for solving everyday business problems. The key is to start small and be specific. Identify one challenge that slows you down or limits growth, whether it’s product recommendations that don’t convert, customers waiting too long for replies, or sales trends buried in spreadsheets.
“Tools like Shopify’s Sidekick make it simple to tackle these challenges without the learning curve. It can analyse store data, surface opportunities, suggest next steps, and even draft customer communications, freeing you to focus on strategy, creativity, and relationships.
“According to Shopify research, nearly half (46%) of business leaders predict AI will help them reach new markets or segments over the next 12 months, but only if it’s put to work. In retail, those who act fast and build momentum with AI now will be the ones shaping customer expectations tomorrow.”
Lisl Pietersz, Solopreneur, Communications and Transition Coach, University of Sydney Business School, AGSM at University of NSW
“Professionals who are not technical and short on time and can still make GenAI their career ally. The secret is knowing how to give the right instructions and easily integrate your chosen AI tool into the task at hand.
“My coaching approach is simple:
First, choose the right AI tool for the job: ChatGPT, Claude, Copilot, Gemini, or MidJourney for images.
Next, keep prompts simple but detailed: state your goal, add context, and say how you want the answer to sound. For example, do you want a funny or professional response?
A good prompt improves accuracy and saves time on revisions, though you’ll still need to refine the response or output. A vague prompt makes AI guess, leading to frustration and wasted effort.
When creating content, write in your own voice, but tell the tool what that is. For example, my style is clear, direct, and grounded.
As an experienced executive coach, I authentically integrate AI-driven tools into most stages of career development so clients can streamline tasks and focus their energy on enhancing communication skills – a key driver of personal and career impact.
Prompt with purpose and passion and the technology becomes an effective career partner.”
Amber Johnson, AI Optimisation Lead, MYOB
“The great thing about AI today is that you don’t need to be an expert to use it or to get real value from it. At MYOB, we think of AI as a teammate you can collaborate with and, if done effectively, it can clear the clutter from your to-do list so you can focus on what matters most.
“In mid 2024, MYOB appointed a GM of AI and Data and created my role as AI Optimisation Lead, reporting to our Chief People Officer. It’s a sign of how seriously we take AI, not just as a tech initiative but as a capability that belongs to everyone.
“That’s why we designed and launched AI Everyday, a program built on mindset, skillset and toolset. It’s about helping every person, no matter their tech comfort level, confidently put AI to work.
“Here are a few things I’ve learned along the way:
- Start with the right use case: AI delivers the most value when it’s solving the right problem. Look for tasks that are repetitive, time-heavy, or could benefit from deeper insights. If the work calls for creativity, analysis, or decision support, AI can be a strong partner. At MYOB, everyone has a goal to re-designingat least two workflows with AI this year and sharing their learnings.
- Use the tools you already have: Many tools you use now, including MYOB, have AI features built in. If you’re new, start simple with AI tools you have access to and get comfortable experimenting.
- Be specific when you ask: The key to getting the most out of your AI friend is being clear about what you want it to do; the clearer your instructions, the better the output.
- Make AI part of the conversation: Put AI in the meeting agenda. Ask team members how they are using it each day and what they would like to experiment with.
- Learn from each other: Our AI champions act as advocates, guiding engagement and sharing AI stories. We also have AI communities on our internal platforms where people share articles, use cases, questions and ideas.
- Track your progress: We run regular pulse checks to see how confidence and skills are growing.
- Know its limits: It is crucial to understand that while AI is a powerful tool, it still has its limitations. It’s important that you always fact-check and cross reference the outputs.
- Use it safely and securely: Protect customer and company data; we’ve set clear safe-use guidelines and training so everyone can use AI and experiment safely.
“When you bring together curiosity, capability, and accessible tools, AI shifts from being a concept to an everyday habit, creating space for meaningful, high-value work.”
Anthony Capano, Regional Director, APAC, Intuit Mailchimp
“You don’t need to be an expert or have hours to spare to get value from AI. The trick is letting it handle the repetitive tasks so you can focus on strategic, higher-value work.
“This could mean automating follow-ups after a purchase, tweaking audience lists for a new product launch, or creating email variations to see which messages resonate. While AI quietly works in the background to analyse data, spot patterns, and offer ideas, and you can pair these insights with your own expertise—enhancing human judgment rather than replacing it.
“Embedding AI and automation into marketing processes, with the help of platforms like Intuit Mailchimp, can enable personalisation at scale and help businesses reach their goals.”
Mark Drasutis, Head of Value, APJ, Amplitude
“You don’t need to be a tech expert to make AI work, you just need to start with a clear, practical goal. Pick a task you want to simplify or speed up, like analysing customer feedback, creating quick reports, or spotting patterns in sales or engagement data.
“Look for AI tools that let you ask questions in plain language and deliver clear, actionable answers, so you’re not buried in dashboards or spreadsheets. Many modern AI solutions can summarise information, highlight key trends, and even suggest next steps, saving you hours of manual work.
“Start small, experiment with one or two use cases, and focus on tools that remove complexity rather than add to it. The aim isn’t to learn all the tech rather it’s to let AI handle the heavy lifting so you can spend your time making better decisions faster.
“With Amplitude AI Agents, we’ve demonstrated the ability to deliver what used to take days and sometimes months to produce down to minutes. The sooner you start experimenting, the sooner you can benefit.”
Monique Biady, Head of Commercial ANZ and Singapore, Checkout.com
“For many ecommerce leaders, the real question isn’t whether to use AI, it’s where it can make the biggest difference to revenue and customer experience, and the checkout is often the best place to start.
“AI can process millions of data points in real time, spotting patterns and anomalies that humans and traditional systems miss. This means it can signal when to block fraudulent activity before it happens, while also recognising legitimate customers who might otherwise be wrongly declined, a problem that quietly costs businesses billions globally.
“False declines are more than just a lost sale, they can damage customer trust and push shoppers to competitors. A payments provider leveraging AI can analyse historical approval and decline data, highlight performance gaps, and automatically adjust decision-making to recover genuine transactions.
“AI-powered tools can also identify seasonal or event-driven patterns, like fraud spikes during major sales periods like tax time, and adapt security measures accordingly. This ensures strong protection without unnecessary friction for trusted customers.
“For leaders short on time, the key is to start with one measurable goal, such as reducing false declines by a specific percentage, and work with a payments partner who can leverage AI without additional technical lift. . With targeted use cases and the right data, AI becomes less about complex technology and more about unlocking revenue, protecting customers, and freeing you to focus on growth.”
Bryan Stallings, Chief Evangelist at Lucid Software
“If you’re not a technical expert, using AI can feel intimidating, but it doesn’t have to. The trick is to focus less on learning complex systems and more on the skills you already use every day. Don’t underestimate the power of soft skills like critical thinking, problem-solving, and clear communication. These aren’t just nice to have; they can make a big difference in how productively you use AI.
“Critical thinking helps you evaluate AI’s responses, spot gaps, and decide when to dig deeper. Problem-solving enables you to frame challenges so that AI can actually help, whether you’re mapping a process, exploring new ideas, or simplifying complex information. Communication skills ensure your prompts are clear and specific, increasing your chances of getting useful results.
“Still, not everyone feels confident using AI. Forty-one per cent of entry-level workers in Australia say they’re not comfortable with it. And while 63 per cent believe training is essential for responsible use, only 14 per cent have received it through work. The good news is you can start building confidence now. Treat AI like a collaborator, not a shortcut. Guide it, question it, and refine it. By pairing it with the soft skills you already have, AI can go from overwhelming to a powerful partner.”
Jarrod Kinchington, Vice President Australia and New Zealand at Smartsheet
“You don’t need to be a tech expert or have hours to spare to start using AI effectively. The real question is: where is your time going? If you’re spending too much of it chasing updates, compiling reports, or repeating the same tasks, AI can take that load off your plate.
“You also don’t need to overhaul your tech stack to see results. Start small. Pick one task you repeat often, like the 30 project status updates you’re asked for each week, for example. AI can draft those instantly from live data, freeing hours for creative and strategic work.
“At Smartsheet, we’ve built intuitive, context-aware AI into the tools you already use, so you can automate workflows, generate insights, assign tasks, and even anticipate risks without coding or knowing exactly where to start. Think of it as a digital teammate: suggesting, acting, and keeping things moving while you focus on what matters most.
“If you’re worried about security, choose AI tools that are model-agnostic with enterprise-grade controls to protect data.
“The future belongs to organisations that make AI accessible to everyone. When every team member can use it easily, AI becomes a strategic advantage. The question isn’t whether you can use AI, it’s what you’ll do with the time it gives you back.”
Elise Balsillie, Head of Thryv Australia and New Zealand
“Small business owners move fast. AI needs to move faster, by quietly connecting the dots across your day so you can serve, sell and stay in control, without shifting gears.
“The smartest AI should not feel like a complex new task. It is the engine underneath the systems can already use – nudging the next action, drafting the messaging, chasing the payment, surfacing the customer who is ready to rebook.
“You do not need to lift AI’s hood to feel its performance. Its value lies in how seamlessly it clears clutter and prioritises progress.
“This is where real business advantage shines, not in bold claims, but in quiet and accurate precision. The quote you did not have to chase, the appointment that confirmed itself and the invaluable insight that landed overnight, while you slept.
“AI provides the helpful rhythm needed for your business With the right AI, your business continues to flow and grow, so you can show up where it counts – with time to strategise, room to grow and tools already working at your pace.”
Victor Horlenko, Head of AI Innovations, Devart
“IIf you’re not technical and short on time, the key to using AI is to start small and focus on practical, everyday wins rather than trying to “learn everything.” You don’t need to be a programmer to get real value from it.
“A good first step is to visit the OpenAI Academy (or similar beginner-friendly resources) to cover the essentials – just enough to understand what AI can do for you. Don’t overthink the theory, focus on examples and quick demos.
“Next, explore Study Mode in ChatGPT. This lets you interact with AI in a guided way, asking questions and testing prompts on the spot. Even five to ten minutes a day is enough to build confidence.
“Start with simple tasks you already do – summarizing emails, drafting documents, generating ideas, or quickly researching a topic. As you get more comfortable, you can try more advanced uses like creating templates, automating repetitive work, or brainstorming business strategies.
“The secret is to make AI part of your routine. The more you use it for real tasks, the faster you’ll discover how it can save you hours, without needing to be “technical” at all.”
Sangeeta Mudnal, Chief Technology Officer, Glu.ai
“For today’s creative professionals, the e-commerce landscape often requires sacrificing authentic expression for technical optimization. Hours are devoted to keyword research, metadata management, and platform-specific formatting, which drain creative energy while delivering diminishing returns. The rise of conversational commerce offers liberation from this technical treadmill.
“AI platforms like Glu.ai aren’t simply another productivity tool but rather the foundation for an evolution in conversational commerce. As an example, Glu’s AI-powered platform helps organize digital assets with automatic tagging, generate tailored content suggestions, and automates time-consuming tasks like bulk cropping and resizing, creating the operational efficiency you need to experiment with emerging AI channels.
“Starting with a solution like Glu means building the creative muscles and operational frameworks quickly and at scale. While other creative producers are still struggling with platform-specific formatting and technical SEO optimization, you’ll be crafting distinctive brand voices that flourish in conversation.”
Caitlin Stephens, Chief of Staff, APAC, Eagle Eye
“Eagle Eye is a company with AI embedded in its product DNA. And, while we are all passionate about AI here, we appreciate that not everyone is highly technical.
“There are a few things organisations can do to help their non-technical teams who are short on time get something out of AI. For us, as well as adopting tech solutions that are tailor-made for office productivity, we’re focused on encouraging experimentation and proactive engagement with AI.
“Firstly, we’ve identified our champions: people with a curiosity for exploring AI solutions who can help lead the way for others who are less technical. AI is moving fast, through our champs we can focus on continuous learning, knowledge-sharing and relatable use-cases that demonstrate the possibilities.
“We’ve encouraged this exploration through an internal hackathon to give people the opportunity to build cool things and show them off. This has not only been fun, we’ve also discovered potential innovations to bring to our day-to-day work.
“We’re staying focussed on our ‘why’. The AI revolution is here, we’re transparent about our vision for AI which will help to enhance our people in the workplace, not replace them.
“You don’t have to overhaul everything to get started with AI. Leveraging the right mix of tools, peer champions, ongoing learning and sharing successes, you can help everyone unlock value from AI.”
Billy Loizou, Area Vice President, APAC, Amperity
“The reliance on manual coding and rules-based systems for preparing customer data has remained a significant bottleneck. Customer data engineering is full of repetitive, painful work, so at Amperity, we built an AI agent called Chuck Data to get rid of it.
“‘Chuck’ is the first AI agent built specifically for customer data engineering. Chuck understands your data and helps you get stuff done faster, whether you’re stitching identities or tagging PII. No orchestration, no UI gymnastics—it’s just fast, contextual, and command-driven work. You tell Chuck what you want to do. Chuck figures out the best way to do it, writes the SQL, and gets it done.
“Chuck Data operates directly within the terminal, providing engineers with the capability to understand, tag, and resolve customer identities swiftly within their Databricks lakehouse. The AI agent facilitates ‘vibe coding’, enabling data engineers to execute complex engineering tasks through natural language prompts.
“The agent applies Amperity’s identity resolution models, which are based on experience with billions of data sets from over 400 enterprise brands, to streamline and optimize common engineering processes.”
Anthony Cipolla, AI Lead, COSOL
“Automation powered by AI presents great opportunities for asset-centric industries. However, organisations across the Australian industry landscape exhibit mixed maturity when it comes to their AI journeys.
“When projects involve multi-million dollar assets and operational environments where failure carries substantial consequences, AI deployment can raise just as many challenges that it promises to overcome. Such challenges concerned factors like cultural differences, expectations around advancement, pace of innovation, maintaining compliance, and educating wider teams.
“Here at COSOL, we adopt the ‘walk, jog, run’ framework, where organisations are encouraged to gradually build their AI capabilities sensibly and safely. This framework sees AI first needing to become trustworthy and repeatable, then later able to deliver real value, before late-stage scaling up into production across the business.
“As organisations embark on new AI initiatives, it's crucial that they create secure boundaries and operate within protected environments. This ensures that sensitive information, whether it’s proprietary data or financial instruments, remains fully safeguarded.
“AI is both a disruptor and an enabler, and no doubt there will be tensions and hurdles along the way. Cultural change, mindset and trust will be key factors that organisations either have faced, are facing or will face along their efforts to modernise with data, AI and automation.”
John Hubbard, CEO, Push Button Webinars
“You don’t need technical skills if you train your AI with your unique IP and expertise. Start by uploading your knowledge base, breaking your process down into core components, such as the biggest pain points, the ramifications of those pains, and the boldest promise you can make about the solution you offer.
“Next, get into some avatar analysis and psychographics so the AI can discover the full profile of your ideal client. Start by capturing the authentic voice of your audience and identifying their fears, aspirations, and internal conflicts. Provide the AI with all of this information, plus some resistance analysis to identify and neutralise potential objections.
“Once your AI is trained on your specific IP, expertise, voice, and writing style along with your ideal audience’s pain points, concerns, desires and needs, it has your practical domain experience to guide and refine its output. This ensures the content it creates will communicate your individual IP into the prospect’s internal conversation better than they could articulate it themselves.
“This saves time by building a repeatable system. Instead of manually creating content from scratch each time, your AI can be instructed on multiple tasks, and it will respond exactly as you would.”
Greg Wilkes, CEO of Develop Coaching
“AI sounds fancy, but for busy builders, it’s just a smarter hammer. You don’t need to be technical just willing to try a tool that frees up your time.
“Start small. Take your last month’s invoices, paste them into a free AI summariser, and instantly find hidden margin leaks. Next, drop your standard project spec into ChatGPT to quickly generate a risk checklist, saving hours of headaches.
“In a recent LinkedIn post, construction coach Greg Wilkes showed how he now quotes straight forward jobs in minutes using AI (using AI to quote complex jobs is only a matter of time). Instead of hours buried in spreadsheets, Greg shows how to simply uploads project details and watches as AI produces accurate, professional quotes almost instantly. The result? More time to win new business and less chance of costly mistakes.
“Finally, if scheduling teams feels like herding cats, plug a simple AI calendar tool into Outlook or Google. Let it juggle your crew schedules while you chase new clients or oversee site quality.
“AI doesn’t replace your expertise – it multiplies it, removing tedious tasks so you can focus on what really matters: building your business. Start with small steps and watch how quickly AI becomes your smartest hire – no tech skills required.
“We are looking forward to your response, should you have any questions please do not hesitate to give me a shout.”
Alon Abraham, Co-Founder and Head of Strategy, TAG
“When people hear ‘AI’, they often think you need to be technical or have hours to spare on learning tools. The truth is, you can get business impact from AI without either. I run a marketing and technology consultancy, so I see first-hand how busy leaders can make AI work for them.
“Start by identifying one or two repetitive, time-consuming tasks you do every week, like creating monthly reports, drafting a digital strategy outline or turning meeting notes into clear action points. Then, delegate those tasks to AI-powered tools.
“The secret is in how you ask. Give context, constraints and examples. Instead of ‘write my monthly report’, try ‘summarise these analytics into a one-page report for executives, highlighting the three biggest wins, three challenges, and clear next steps’. The better the brief, the better the output.
“Build a prompt library so you’re not starting from scratch every time. Keep it somewhere easy to copy and paste. Small, targeted use cases turn AI from hype into a genuine time-saver.
“I also use ChatGPT’s voice feature when walking or in the car – it’s like having a quick brainstorm with an on-demand strategist.”
Maria Kathopoulis, CEO & Chief Marketing Officer at UNTMD Media
“You don’t need to be a coder in a hoodie to make AI work for you. You just need to know where it plugs into your day. Think of AI as your over-caffeinated assistant who works 24/7, never takes a sick day, and can read the internet in seconds.
“Start small, win big:
- Content drafting Tools like ChatGPT or Jasper can write your emails, ad copy, or LinkedIn posts in minutes. Your job? Add the human polish. And remember the output is all about how good the input is.
- Meeting notes: Read.ai, Gemini, Zoom will transcribe, summarise, and pull out action items faster than you can say “next steps.”
- Email triage: Superhuman AI sorts your inbox, flags urgent emails, and even suggests replies.
- Customer insights: Klaviyo AI or HubSpot predictive scoring tells you which leads are most likely to convert so you can stop guessing.
“Rule of thumb: Don’t start with “AI strategy.” Start with one mundane, repetitive task and outsource it to AI. When you see hours coming back into your week, add another. The goal isn’t to learn AI. It’s to make AI learn you, so that your time goes where it actually moves the needle.”
Julie Lawrie, CEO / Co-Founder, Amplifyo
“The hype around AI continues, but the day-to-day work pressures on time-poor SMEs with varying tech expertise, might mean you’re yet to utilise AI tools in your role. The gap between AI adopters and those who are yet to fully engage might seem to be growing.
“The reality however is not as bleak. The good news is that the pace of change in AI, and the many variations like generative AI, agentic AI, and automation AI equals better accessibility and quality for new users today.
“So…where do you get started? Our founders have worked with SMEs for a decade to streamline growth and now use AI in the leanest way – especially teams with varying tech-comfort. Here’s where we recommend starting:
- List your daily tasks
- Circle any tasks that you think could possibly be improved with automation
- Start a simple Google search for “AI for [insert task names]” and see if there are options that look good
- Dedicate 30-minutes to test the options (most good AI tools come with a free trial period, onboarding instructions and videos, and the chance to play)
“Start with task & productivity questions….it will make it easier to use AI.”
Tracy Ford, Human Resources Consultant, Concept HR Services
“Many SME leaders I speak with are curious about AI, but may put it lower on their priority list, often because it feels overwhelming.
“You don’t have to overhaul your business to test AI.
- Use AI built into applications you already know, such as Microsoft 365 with Copilot or Google Workspace with Gemini.
- Stick with one AI application at first so it learns about you and gives better results.
- Follow one or two AI newsletters or thought leaders for ideas and ready-to- use prompts.
- Speak to it in everyday language and let it handle the technical part.
- Begin with a topic you know well so you can easily spot errors.
- Voice record a short instruction on your phone while you’re out and have the output ready when you return.
- Ask it to identify tasks you could delegate, freeing time for higher-value work.
- Use it to outline a change management plan for introducing AI into your business.
- Use trusted tools, protect data, and check outputs so AI becomes an asset, not a liability.
“Setting aside 10-15 minutes a day to experiment and learn will put you in the best position to decide if, when, and how AI will add value to your business.”
Peter Curran, Founder & Business Development Manager, Digital Surfer
“We’ve all heard those stories about the company who replaced their admin, marketing or whatever team with an AI system and now they’re generating big $$$. Want to know what they’re doing? Yes, they have an AI system, but they also have a very talented human who knows their stuff in that area, whether marketing, admin or whatever, who is programming these tools and driving those results. They likely actually have a whole team of these incredible humans using AI to speed up their results, using it to do more, not necessarily to replace.
“Now, how can you do that? I’ll be blunt, if you don’t already know how to do something and can’t clearly communicate to the AI tool what you need, you’re not going to get the same results.
“It’s not to say you can’t use AI to do great things, but you need to understand your competitors are probably using it too, and everyone is starting to sound the same because of it. You need that human element to really push that generic limitation we’re still getting from AI.
“BUT if what you need is to just have something, anything, up on your socials, website, chat bots, to give you a basic idea of what to do or to give you bullet points from a report, go from it. Just keep in mind it’s not magic.”
Zanda Wilson, Head of Corporate Communications, Sound Story
“You don’t need to be technical or have endless hours to benefit from AI at work. In PR and communications, the arrival of AI has sparked debate, as any disruptive technology does, but sticking to the old playbook is no longer an option in 2025.
“Tools like ChatGPT are already replacing traditional search for many people, making them valuable for client research. Reviewing recent media coverage of a company not only deepens your understanding of their business, it also reveals what the public sees. Using AI when pressed for time is not the same as using it out of laziness. Journalists can often spot a media release generated by AI, whether full or in part.
“The deeper advantages come from refining your skills with a platform, spending a few minutes each day improving your prompts and even creating custom GPT models tailored to each client. Embracing these efficiencies ultimately allows more time to be spent on areas where AI can’t help, such as deep strategy work.
“With a growing number of free and paid courses available to master the basics, there’s little barrier to building AI into your workflow in a way that is both efficient and authentic.”
Seamus Phan, CTO, McGallen & Bolden Pte Ltd
“For businesses using AI for customer service, RAG (retrieval augmented generation) are AI chatbots that are trained with your own documents and are very easy to set up and maintain. Public chatbots may “hallucinate” since they were fed on public data, which can be biased or incorrect. With RAG, your chatbot will only be trained with your own information and can be constrained with guided prompts. I suggest that public-facing RAG chatbots be trained with public domain information only, including your website, blog, brochures, spoken video, and public documentation.
“Beyond public-facing RAG chatbots, emerging businesses can consider using offline RAG that runs off their own devices, such as desktops and laptops, available on Linux, macOS, and Windows. These bots will not leak your internal information and have no telemetry. Since they run locally, these bots are safer to use with your own proprietary internal documents for your own analysis and queries. For example, you could use them to query your financial and sales spreadsheets or text documents.
“It is wise to bootstrap with AI adoption and to be prudent in expenditure and to ensure that the right documents meet the right tools to constrain the data privacy and IP protection.”
Rolf Howard, Managing Partner, Owen Hodge Lawyers
“AI tools are a great way to save time and streamline tasks, even for non-technical users. However, to use AI well, it must be used responsibly.
“Firstly, when inputting data into AI, be cautious with confidential or personal information. Never use actual customer data to train an AI model. Many AI platforms have terms of service that allow them to use your input to refine their models, so you must select the appropriate settings to prevent this.
“Secondly, always exercise human oversight over the AI’s output. The technology can “hallucinate” or generate incorrect and even biased information. Always review and verify the AI’s generated information for accuracy before using it. This is your responsibility, as the law on liability for AI-generated errors is still evolving.
“Finally, be aware of copyright issues. Since AI output may not meet the requirements for copyright protection, you likely won’t have legal ownership rights to it. If you want to copyright the information, you must add independent human skill or talent to the final product to make it a discernable part of the work.
“Ultimately, it’s crucial to not let AI’s extraordinary potential distract you from taking a responsible approach to the use of AI.”
Discover Let’s Talk Business Topics
Keep up to date with our stories on LinkedIn, Twitter, Facebook and Instagram.
