Dr. Laurence Lock Lee

Thanks to Kirsten (KP) Patterson, David Downs and Anna Campbell from the Institute of Directors in New Zealand on their reflections and lessons for #governance #directors from the Spark AI* Mini-MBA. Here is my favorite thought on how we should use #generative AI: Treat #AI like a 'high performing graduate'. It has a good base level of knowlege, needs adult supervision in the office, needs to be fact checked, needs good context and requests, your communication will determine it's responses and it will not always perform consistently. Let me know your thoughts on this take. Here are some more of my main takeaways: We have previously talked about an extending digital divide, and we may be seeing an AI class of leaders developing, with some being left behind. As Generative AI is new to a lot of people, many feel they are 'lurkers' - interested but not engaging or not knowing how to engage. You have permission to engage with AI, there is no one that is going to give you an endorsement or permission to explore and learn. Most directors and management are learning about AI together, at the same time. There is room to be an interested learner. Being a learner is key cornerstone of ongoing good governance. There is value in using AI as a "strategy companion", moving past the gimmicky uses of AI that most people engage with. For example in the images there is a clear divide in the depth of use of AI. Understanding the protection of sensitive and private information, AI should be looked at as a significant productivity tool for organizations. David shared a matrix of different uses: - Low risk / Low effort: Email drafting and responses Basic note taking Simple task management Information retrieval - Low effort / High risk: Basic sentiment analysis on public opinion Automated document reviews - Low risk / High effort: Personalised learning and development Advanced meeting summaries Automated report generation Smart scheduling assistance - High effort / High risk: Strategic planning ideas Comprehensive personal knowledge management Enhanced decision support Intelligent virtual assistance AI can help creativity, stimulating and prompting ideas. It can also help with more mundane decisions, allowing space for more intentional thought of complex issues. AI is often thought of as a technological tool, but this can be a limiter on how we can use it, personally as leaders and professionally as organizations. You can start prompts with a caveat, 'I want facts, validate and quote the source of information, and no generated or hallucinated information'. You can also also add these sorts of directions into ChatGPT settings, have a look in the cog icon. This can be a brilliant tool for neuro diverse people, as it can take a large amount of verbal information, and summarize this. There are some things that AI will be an expert on, such as getting AI to write a draft policy on AI use. *I have used AI in this post to refer to Generative AI.

20

10 Comments

Do you want to know if an article, white paper or resource was written or assisted by AI? The Institute of Directors in New Zealand has committed to placing this symbol on their AI assisted resources. Some AI image creation apps have watermarks or logo's to denote it's source. And recently newsrooms have been found to be using #AI and had to wrestle with their standards. I am not interested in having these AI warnings, as I will judge the brand attached or promoting the work. I trust Kirsten and the Institute of Directors team to put out good materials, and that that extends to anything they create with AI and chose to publish. If the quality is poor because of AI, I will lose trust in their brand. The same with the New Zealand Herald, if they promote or publish a poor quality article because of AI, I will judge their brand and trustworthiness. I see real value in images and video having watermarks which spread differently than text, but I will always look to the publisher, agent or organization pushing an article. The good or poor AI will still rely on their quality assurance and standards. What are your thoughts here? Do you like or find these sorts of triggers helpful?

11

3 Comments

Generative AI is "kind of dumbing down our language," Te Reo Māori language and culture educator and author Te Mihinga Komene explains in the latest episode of World Privacy Forum's Privacy on the Ground podcast. She and others in her community have provided their knowledge and contributed countless hours over the years to projects operated by several tech companies. Some of that work has been done on an entirely volunteer basis without pay, but always in the recognition that the products and services in development could affect the future of their language, culture and community. In our talk for the podcast, she shared some of those stories - positive and negative. I was lucky enough to speak with and spend time with Te Mihinga Komene while in Rio de Janeiro, Brazil in June for the ACM FAccT conference, an important annual conference that brings together scholars and practitioners from around the world who are interested in fairness, accountability, and transparency in socio-technical systems including AI and machine learning. Komene was at FAccT to speak alongside several other Indigenous language experts about efforts to shift tech and AI-related language research, data collection and use toward more “responsible non-extractive” approaches involving collaboration with Indigenous communities that center “language communities in [Natural Language Processing] ethics, fairness and accountability.” During that session and later in her podcast talk, she described specific examples of power imbalances in AI research and development that she and her community have experienced – including an ironic yet poignant example that was unintentionally illustrated during the session. At the time, Komene had recently shifted the focus of her research as a PhD student -- at University of Otago (Te Whare Wānanga o Ōtākou) -- toward generative AI. But her transition was not about keeping up with trends. This was an existential one, a recognition of an already emerging reality: that generative AI systems could drastically alter the ways Te Reo Māori is communicated. If you're interested in Te Reo #Māori language and culture, Indigenous culture and #Indigenous data sovereignty, #generativeAI, NLP, language data used in AI, -- or of course, data #privacy -- this talk and the other episodes in the Privacy on the Ground podcast are well worth the listen. Put this one on your holiday listening list! Apple: https://lnkd.in/gyvzrNdn Spotify: https://lnkd.in/gj9XHT9g iHeart: https://lnkd.in/gupEvygp

14

3 Comments

Important information on making meaningful apologies.

1

I was happily surprised to see this investigation about abuse of the tendering process in Innovation Aus today. I would love to know how the tendering process is supposed to work - because I am confident that it is rarely working in this way. One of the many challenges in finding suppliers - in most Technology consulting, and especially in the CRM space, where I focus - is getting to a point where the requirements can be accurately described. When this is not done well, one side loses, and often badly. As many organisations do not have the skills to put together a project brief, it is natural to reach out to a connection for assistance. If this connection then tenders for the project, they undoubtedly have an advantage. And, it requires a huge amount of time and effort to put together a response to a tender - so an organisation is unlikely to make that investment unless they have a real prospect of winning the work. There are also disadvantages to the tenderer in using an unknown supplier. I feel strongly that identifying the best team to do a job is achieved by a gradual back-and-forth process where each side gets to know the other and the detail of the work is thrashed out at the same time. As the tender process makes this impossible it is hardly surprising that people find ways around it. Recently, I have been involved in a tender that was in two phases - phase one was purely design of a CRM solution, and phase two was the implementation of that design. Any organisation who tendered for phase one was excluded from phase two. This sounds good in theory, but I struggle to see how the implementation can be successful if no-one involved in the design is still around. All in all, tenders are tough - for everyone involved in them.

3

It's refreshing to see how many people are responding to our fortnightly FREE webinars, where we show the practical side of applying good #governance across the organisation (regardless of its jurisdiction, profit, non-profit and state entities). As more #directors, governance, risk, compliance specialists, company secretaries and auditors attend these webinars from all parts of the world, a few things have become very clear... Key Take-aways are: 1. Take the fear out of this critical topic of governance 2. Make it as simple as possible to understand and address 3. Show the HOW and WHAT factors within the organisation's #governanceframework 4. Provide the value behind good governance and ROI impact 5. Show the quick wins If governance is not properly defined and understood across the organisation, if governance is not measured, it cannot be managed and this will overwhelm leaders and their organisation. It's that simple... Once the fear is removed and the ease of a practical governance application is provided through seamless technology, leaders are better equipped to run and grow their businesses!

3

Nice to see the Australian Press Council upheld Yoti’s complaint that Crikey’s claim it had fooled Yoti’s facial #ageestimation was “significantly misleading” & have forced Crikey to issue a correction to its readers. The APC concluded Crikey had breached General Principles of Prctice 1, 2, 3 & 4. Yoti understands that for a little while longer some journalists & some of the public may assume that facial age estimation could be tricked using photos & age filters but it’s not credible in 2024 for tech journalists or some self asserted trust & safety experts to ignore the fact that leading AV providers perform liveness checks when capturing live faces for age estimation (or face matching). Yoti presentation attack detection (PAD) & injection attack detection (IAD) models are very hard to spoof. In the last 8 months that we have been running an independent HackerOne $1,000 bounty scheme, just 3 ethical hackers have beaten our patented Secure Image Capture and Processing (SICAP) IAD model. But some evidence submitted to the US Supreme Court regarding the Texas AV for porn law case claims FAE is easy to spoof ! We also offer businesses the choice to use alternative world class liveness including FaceTec, Inc. & ID R&D & businesses can choose other world class liveness from the likes of iProov. Whilst iBeta Quality Assurance have certified dozens of liveness vendors against US National Institute of Standards and Technology (NIST) for Level 1 & Level 2, there has not been widespread independent testing of injection attack detection despite the explosion in deepfakes. The well funded, high profile Australian Age Assurance Trials in Feb to Apr 25 will test PAD & IAD when used for face capture for age estimation & ID Doc verification after which it will be very hard for opponents of some age check methods to claim they can’t be effective. You can’t buck science for too long.

46

1 Comment

🇦🇺 AI Impact Navigator: A Guide for Responsible AI Implementation - The Australian Government Department of Industry, Science and Resources, National Artificial Intelligence Centre and CSIRO have released a new guide to help businesses navigate the impacts of artificial intelligence (AI). The AI Impact Navigator is a framework for companies to use in assessing and measuring the impact and outcomes of their use of AI systems. It provides a way for company leaders to communicate and discuss what's working, what they've learned, and what their AI impact is. The Navigator is structured around 4 dimensions to help you leverage AI for positive impact: 💠 Effective AI and community impact   💠 Customer experience and consumer rights   💠 Workforce and productivity   💠 Social licence and corporate transparency   🔗 Find more information here 👉 https://lnkd.in/gN3DbkAs

29

1 Comment

A recent analysis of two major transformations showed just how critical it is to have the right enablers in play for a successful outcome. Tony Boyd’s article in the AFR (https://ow.ly/ppZb50RqjTH) about two such projects illustrates just how critical leadership, planning, and the right level of engagement with third parties is for the outcome of transformation initiatives. Visionary leadership, strategic planning, meaningful engagement with external experts, and a keen focus on outcomes form the backbone of successful digital transformations. For leaders embarking on high-risk projects goes beyond oversight; it's about integrating change into the very fabric of the organisation, ensuring alignment with the overarching vision. Third-party consultants play their role best as enablers, not as the driving force, supporting internal capabilities with their expertise. Remember, accountability for the transformation's success rests firmly with the executive leaders. When leadership remains hands-on, the trajectory of core projects aligns closely with success. For more insights: https://ow.ly/B0F450RqjTG Follow the Bulletin: https://ow.ly/96aH50RqjZ #DigitalTransformation #Leadership #ProjectManagement #ITGovernance #CorePlatforms #StrategicPlanning #AccountabilityInLeadership

2

Given AI is challenging everything in the digital space, are Australian executives more informed or hands on? From a governance perspective it raises questions about Directors making decisions on AI.

5

Thanks to Dan Te Whenua Walker (Microsoft Global Partner Solutions), 🎙Hilary (Hils) Walton (Microsoft Technology Strategist) and Anthony McMahon (Technology Strategist, Target State) for supporting Community Governance Aotearoa's Board Talks. Thanks Rose, another good one! One of the most interesting takeaways for me was the resources that Hilary referenced, including the following: https://lnkd.in/gGEe-QFG Generative AI use has nearly doubled in the last six months, with 75% of global knowledge workers now using it. 46% of users started using AI less than six months ago. Employees grappling with the pace and volume of work, are even bringing their own AI tools to work. This is creating shadow infrastructure which organizations and leaders need to be aware of and prepare for. Essentially if your policy (if you even have a policy) limits staff from inputting company data or having AI tools on work computers, this is not stopping them using AI on other devices. While this keeps some material secure, it isn't stopping the use of AI, and has the impact of it being effectively invisible to leaders. And here's the challenge - while leaders agree AI is essential, many organizations lack a clear plan to move from individual impact to driving the bottom line with AI. This report looks at: Immediate ROI pressure - Leaders are hesitant, despite AI inevitability, due to the pressure to show immediate return on investment. Business Transformation - Just like the internet or the PC, business transformation requires broad adoption. Organizations that apply AI strategically will pull ahead. Talent Shortage - Despite fears of job loss, leaders report a talent shortage for key AI roles. Career Impact - For employees, AI raises the bar but can also break the career ceiling. You can find the full report details here: https://lnkd.in/gGEe-QFG The webinar recording, resources and other Board Talks can be found here: https://lnkd.in/gY7GEQUq #governance #boards #directors #AI

11

7 Comments

Deep fake videos create disinformation that can deceive parts of society as evidenced by this example of fake Tom Cruise denigrating the Olympics. At some point, nefarious actors are likely to weaponise fake videos against companies in NZ too. Brand protection takes on a whole new meaning requiring new considerations. https://lnkd.in/gTAmt7tf #businessnz #misinformation #disinformation #businesssecurity #brandmanagement

6

Speaking at the Australian Public Service Commission's Mastercraft Series today on 'Trust in Government's use of AI'. While responsible and safe use of GenAI is essential, I will show that there is an opportunity cost here in Australia to being too risk averse. Rather than focusing on IT investments and system-level change as the entry point, there is immediate benefit that comes from enabling the APS - and all workforces - to integrate GenAI into the daily workflows of all workers.

10

The Australian Financial Review | Tom Burton Three PhD economists, led by Stanford University Professor Nick Bloom, created two work groups of 800 each in the large Chinese web firm Trip.com. One group worked from the office full-time , while the other worked at home twice weekly. The Stanford research decisively revealed that employees who work from home two days a week are just as productive, likely to get promoted and far less prone to quit than those who work full-time in the office . Resignations fell by a third among workers who shifted from working full-time in the office to a hybrid schedule . Women, non-managers and employees with long commutes were the least likely to quit their jobs when their treks to the office were cut to three days a week… #SDGs #HybridWorkplace #WFH #WFA #WalkableUrbanism #20MinuteNeighbourhoods #TimelessUrbanism #IMBY #ADU #EBikes

15

3 Comments

As we kick off an Aotearoa-wide #data #governance (DG) initiative, a reminder of this piece I presented for Canterbury Tech last year, and a reminder that DG is relevant in _every_ sector and industry, and to all of us :) Lemme know if I can help you and yours peeps, too! https://lnkd.in/gQ-AwMN7

4

Fantastic to see the Australian Institute of Company Directors leading the way with a Directors Guide to Artificial Intelligence being released today. AI is such a widely discussed and debated topic at the moment and it's great to have some clear guidance on how Directors can best ensure that the benefits of AI are harnessed whilst ensuring that appropriate guardrails are in place. It’s great information generally even if you aren’t a Director. The release includes an Introduction to AI, a Directors Guide to Governance of AI plus an SME and NFP checklist which will be really useful. Looking forward to getting up to speed on this content. #artificialintelligence #governance

15

1 Comment

TOO MUCH POLICY, NOT ENOUGH ACTION? Very pleased to see that Tech Policy Design Centre (TPDC) will keep going, despite being kicked out to fend for itself a little earlier than expected. Congrats Johanna Weaver. But, is that an elephant I can see in the room? We have the Tech Council of Australia, AIIA, ACS (Australian Computer Society), TPDC and more, all offering comment and guidance on technology policy, and yet, Australia is STILL rapidly falling in to a resources and agriculture driven banana republic-esque abyss. An optimist might look at this and say thank god these groups exist. Imagine how woeful our governments would be with out them. A pessimist might suggest that if one is to judge based on results, this policy could even be making things worse! I'll give the policy-meisters the benefit of the doubt and thank them for avoiding a complete calamity, but either way, the results are the results and that means despite their best efforts, policy alone just ain't cutting it. If we're going to turn this ship around, we need a new approach to the tech sector in Australia. Follow SOFTWARE Australia today and help us build a software industry to take on the world. https://lnkd.in/d55eaQDW

7