Whose side is AI on?

At the risk of coming across as some sort of Luddite, recent commentary on Artificial Intelligence suggests that it is only natural to have concerns and misgivings about its rapid development and widespread deployment. Of course, at its heart, it’s just another technology at our disposal – but by its very definition, generative AI is not passive, and is likely to impact all areas of our life, whether we invite it in or not.

Over the next few weeks, I will be discussing some non-technical themes relating to AI – creativity and AI, legal implications of AI, and form over substance when it comes to AI itself.

To start with, these are a few of the questions that I have been mulling over:

– Is AI working for us, as a tool that we control and manage?  Or is AI working with us, in a partnership of equals? Or, more likely, is AI working against us, in the sense that it is happening to us, whether we like it or not, let alone whether we are actually aware of it?

– Is AI being wielded by a bunch of tech bros, who feed it with all their own prejudices, unconscious bias and cognitive limitations?

– Who decides what the Large Language Models (LLMs) that power AI are trained on?

– How does AI get permission to create derived content from our own Intellectual Property? Even if our content is on the web, being “publicly available” is not the same as “in the public domain”

– Who is responsible for what AI publishes, and are AI agents accountable for their actions? In the event of false, incorrect, misleading or inappropriate content created by AI, how do we get to clarify the record, or seek a right of reply?

– Why are AI tools adding increased caveats? (“This is not financial advice, this is not to be relied on in a court of law, this is only based on information available as at a certain point in time, this is not a recommendation, etc.”) And is this only going to increase, as in the recent example of changes to Google’s AI-generated search results? (But really, do we need to be told that eating rocks or adding glue to pizza are bad ideas?)

– From my own experience, tools like Chat GPT return “deliberate” factual errors. Why? Is it to keep us on our toes (“Gotcha!”)? Is it to use our responses (or lack thereof) to train the model to be more accurate? Is it to underline the caveat emptor principle (“What, you relied on Otter to write your college essay? What were you thinking?”). Or is it to counter plagiarism (“You could only have got that false information from our AI engine”). If you think the latter is far-fetched, I refer you to the notion of “trap streets” in maps and directories.

– Should AI tools contain better attribution (sources and acknowledgments) in their results? Should they disclose the list of “ingredients” used (like food labelling?) Should they provide verifiable citations for their references? (It’s an idea that is gaining some attention.)

– Finally, the increased use of cloud-based services and crowd-sourced content (not just in AI tools) means that there is the potential for overreach when it comes to end user licensing agreements by ChatGPT, Otter, Adobe Firefly, Gemini, Midjourney etc. Only recently, Adobe had to clarify latest changes to their service agreement, in response to some social media criticism.

Next week: AI and the Human Factor

Perfect Days – and the Analogue Life

Last week I watched “Perfect Days”, Wim Wenders’ lyrical film about a gentle soul who diligently goes about his daily labour accompanied by a soundtrack of classic songs. Most of the featured music is 50-60 years old, and all of it heard via cassette tapes – no radio stations or internet streaming services were harmed in the making of this film!

Not only does our hero cling to cassettes, we never see him use the internet, e-mail or a smart phone. We don’t even know how he accesses his money – presumably he gets a weekly wage packet containing cash, so no need to visit an ATM or pay with a credit card. To cap it all, he doesn’t own a TV, and his hobbies include reading second hand paperback books, taking photos with a 35mm film camera, and cultivating plants from cuttings he finds in the course of his daily routine.

We don’t really need to know his backstory, although we get the occasional glimpse. What we are presented with is someone who is living an outwardly simple life, almost exclusively analogue, and with very little technology involved. (In fact, the public toilets he cleans for a living are far more hi-tech than anything in his personal world.) I suspect for many people, our empathy for the character’s disposition may easily become envy at how stripped down and uncluttered a life he leads. The fact that he doesn’t appear to have any family or other obligations (and doesn’t have to spend hours in pointless team meetings or on endless Zoom call) no doubt help facilitate this state of being – yet we suspect there is a lot going on in the inside.

But it is certainly a parable in favour of all things analogue.

In fact, as I write this I am listening to a recent album by Tarotplane on a cassette player. He is one of many contemporary musicians who choose to release their work in this format, and along with the recent vinyl revival, they are helping to keep analogue alive. It’s a trend we can see in events like Record Store Day (and it’s younger sibling, Cassette Store Day), books by Damon Krukowski and Robert Hassan, and symbolic vinyl moments in recent film and TV shows such as “Leave the World Behind” and “Ripley”. In the former, the absence of internet and streaming brings a turntable into play; in the latter, a clutch of 7″ records (in picture sleeves!) are among the few possessions the eponymous hero chooses to take with him. Elsewhere, Lomography continues to find new fans of film photography, and on a recent visit to Hong Kong, I was surprised at the huge display of Polaroid cameras and film at Log-On department store.

Not all this fascination with analogue is about nostalgia, fashion, or fadism (or even fetishism). In some quarters, people are becoming concerned that their favourite films, TV programmes, music and video games may disappear from hosting services and streaming platforms, or their cloud storage may get wiped. So they are keeping analogue versions and hard copies as a back-up.

Finally, and picking up a thread from “Perfect Days” itself, I’m not entirely convinced that a 1975 Patti Smith cassette is worth $100, but I do own an original copy of a very rare cassette that has sold for as much as $180… probably because it has never been reissued, is not available to stream or download, and is a great example of early, DIY electronic music made on basic synths in the early 1980s. You couldn’t imagine an mp3 ever commanding that sort of price, unless it was in the form of an NFT, of course.

Next week: False Economies – if it’s cheap, there must be a reason!

Pitch X’s Winter Solstice

The latest Pitch X event, organised by Academy Xi and hosted by YBF Ventures, was held a few days before the (Southern hemisphere) mid-winter – there may not have been any mulled wine, but there was still a warm atmosphere on a cold and wet Melbourne evening. The judging panel was drawn from YBF, Melbourne Angels, Linfox and Clearpoint Ventures.

The usual format applied: 11 startups were each given 90 seconds to pitch, followed by a 90 second Q&A with the judges. The top three were then brought back for a 5-minute pitch, and 4 minutes of Q&A.

The pitches, in order of presentation were (links in the names where available):

Startup 101

A self-styled online startup school, targeting university students and recent graduates. The core premise is that entrepreneurship is not being taught to undergraduates. The judges asked about the MVP, which was not clear, nor was there a breakeven forecast based on the number of students. The founder is offering a freemium model, based on memberships and services. Looking $500k for software development and marketing in China (a key demographic for this business).

Studio Ninja

This is a cloud and mobile PaaS solution for professional photographers. I first covered Studio Ninja in late 2016, when they pitched at a StartupVic event – and it’s great to see that they have managed to bootstrap themselves this far.

Professional photography is competitive, but margins are low. Studio Ninja offers an end-to-end platform for scheduling, contracts, payments etc. They have now integrated with Xero, QuickBoooks, Google, PayPal and Stripe, and have built a community via their chat app, Facebook group and Instagram account. At a basic $29.95 per month, they now have 4,000 paying subscribers, mainly in Australia, UK and US. but need to reinvest in product development, scaling and building further efficiencies. Users are offered a 30-day free trial, with an average 25% conversion rate, thanks to the hook of discounts for early sign-ups, plus a referral program.

RoamingDuck

Calling itself the “Uber of travel”, RoamingDuck offers travelers access to curated itineraries, based on their personal preferences. Using freelance resources (along the lines of Upwork and Airtaskr), the service uses a travel planning dashboard on which the customer and the curator can collaborate. With a quick turnaround, RoamingDuck can help customers build and review an itinerary within 12 hours. With the ability to consolidate and share, the content is easily accessible to users, who can plan anything – even supporting “self-plan” users with a search function.

Freelance curators come from the ranks of existing travel bloggers and services like Travelo, and are subject to a vetting process. There is also an escrow system, so freelances only get paid when the customer is satisfied. Normal travel agents are quite restricted on what they can access or offer, and services like Skyscanner are great for searching individual fares – RoamingDuck is solving the planning issue, and building the itinerary. Asked whether RoamingDuck can support actual bookings, the founder will likely implement this via APIs.

Wastr

According to the founder, households waste about 20% of the food they buy. Wastr is an AI-powered app that is designed to help consumers use what they purchase, rather than letting to go to waste. The solution allows subscribers to scan their grocery receipts, and in return they will receive recipes, notifications on expiry dates, plus other reminders. The app is offered under a freemium model, with a paid service starting at $2 per month.

VRWalker

Described as “VR for your foot” (or the “mouse” for VR), this is a motorised shoe device that allows wearers to experience”walking” within VR applications, without actually moving.
It’s an idea that has been around for a while, but the founder claims to have filed key patents. The shoes work on the concept of intuitive locomotion (linked to the dantian, or our centre of gravity), and are intended to be much cheaper and much more convenient than existing treadmill-based solutions. The founder hopes to have a working prototype by the end of this year. Likely customers will come from areas like construction, engineering and gaming. The judges asked about how insurance would be handled, and the device could be bundled with existing VR headset devices.

The Nurture Project

The Nurture Project is designed to teach life skills to deal with anxiety issues, which according to the founder, affect 30% of the population. Unlike other solutions, this treats the causes as well as the symptoms, using a well-being model built on 5 core pillars, and delivered via a 12-week online program. It is currently aimed at women in their 30s and 40s.

Natural MedTech

Designed to boost the immune, hormone and nervous system naturally, this has come out of a CSIRO project, with a scientific basis that has been peer-reviewed.

Magicast

This is a decentralised online podcast recording and editing service. Existing podcast software is either too complex, or too expensive. Instead, Magicast uses web-based programme development, publication and distribution, offering a two-sided marketplace for content, sound effects, music etc. The judges asked about international competitors, given that podcasting is very much a cottage industry, with relatively few barriers to entry.

Turtle

Something akin to an Uber courier service, Turtle enables customers to obtain goods from overseas that are not available where they live. Targeting expat and diaspora populations, the platform has an escrow function to provide a level of trust. It was not clear who would be responsible for tax, customs and quarantine issues.

Young Shaman Foundation

Having run a number of leadership development retreats on country for women in indigenous communities, the founder is now seeking funding to develop and extend the program she currently offers.

SecureStack

Helping companies to secure the cloud, with a focus on cyber security, the founder pointed out that key problems are caused by “cloud sprawl” – the uncontrolled proliferation of content, services and applications hosted and running on cloud-based servers. Using a proprietary cloud infrastructure security design, the startup has already secured two clients and $100k in revenues. Now looking to raise $2m, for an 18-month runway, in order to gain 100 clients. The solution is agnostic as to which cloud service clients use. Traditional cloud management and compliance is saturated, whereas SecureStack’s value proposition is in the security layers.

After much deliberation, the winners were:

1. StudioNinja
2. RoamingDuck
3. VR Walker

Next week: The Metaphorical Glass Jaw

The Ongoing Productivity Debate

In my previous blog, I mentioned that productivity in Australia remains sluggish. There are various ideas as to why, and what we could do to improve performance. There are suggestions that traditional productivity analysis may track the wrong thing(s) – for example, output should not simply be measured against input hours, especially in light of technology advances such as cloud computing, AI, machine learning and AR/VR. There are even suggestions that rather than working a 5-day week (or longer), a four-day working week may actually result in better productivity outcomes – a situation we may be forced to embrace with increased automation.

Image Source: Wikimedia Commons

It’s been a number of years since I worked for a large organisation, but I get the sense that employees are still largely monitored by the number of hours they are “present” – i.e., on site, in the office, or logged in to the network. But I think we worked out some time ago that merely “turning up” is not a reliable measure of individual contribution, output or efficiency.

No doubt, the rhythm of the working day has changed – the “clock on/clock off” pattern is not what it was even when I first joined the workforce, where we still had strict core minimum hours (albeit with flexi-time and overtime).  So although many employees may feel like they are working longer hours (especially in the “always on” environment of e-mail, smart phones and remote working), I’m not sure how many of them would say they are working at optimum capacity or maximum efficiency.

For example, the amount of time employees spend on social media (the new smoko?) should not be ignored as a contributory factor in the lack of productivity gains. Yes, I know there are arguments for saying that giving employees access to Facebook et al can be beneficial in terms of research, training and development, networking, connecting with prospective customers and suppliers, and informally advocating for the companies they work for; plus, personal time spent on social media and the internet (e.g., booking a holiday) while at work may mean taking less actual time out of the office.

But let’s try to put this into perspective. With the amount of workplace technology employees have access to (plus the lowering costs of that technology), why are we still not experiencing corresponding productivity gains?

The first problem is poor deployment of that technology. How many times have you spoken to a call centre, only to be told “the system is slow today”, or worse, “the system won’t let me do that”? The second problem is poor training on the technology – if employees don’t have enough of a core understanding of the software and applications they are expected to use (I don’t even mean we all need to be coders or programmers – although they are core skills everyone will need to have in future), how will they be able to make best use of that technology? The third problem is poor alignment of technology – whether caused by legacy systems, so-called tech debt, or simply systems that do not talk to one another. I recently spent over 2 hours at my local bank trying to open a new term deposit – even though I have been a customer of the bank for more than 15 years, and have multiple products and accounts with this bank, I was told this particular product still runs on a standalone DOS platform, and the back-end is not integrated into the other customer information and account management platforms.

Finally, don’t get me started about the NBN, possibly one of the main hurdles to increased productivity for SMEs, freelancers and remote workers. In my inner-city area of Melbourne, I’ve now been told that I won’t be able to access NBN for at least another 15-18 months – much, much, much later than the original announcements. Meanwhile, since NBN launched, my neighbourhood has experienced higher density dwellings, more people working from home, more streaming and on-demand services, and more tech companies moving into the area. So legacy ADSL is being choked, and there is no improvement to existing infrastructure pending the NBN. It feels like I am in a Catch 22, and that the NBN has been over-sold, based on the feedback I read on social media and elsewhere. I’ve just come back from 2 weeks’ holiday in the South Island of New Zealand, and despite staying in some fairly remote areas, I generally enjoyed much faster internet than I get at home in Melbourne.

Next week: Startup Vic’s Impact Pitch Night