Market hype and the growing popularity of artificial intelligence has pushed companies to introduce the new technology into their product strategy. It has resulted into a growing presence of AI in almost every new product and service. Companies are extensively exploring AI and machine learning as a part of their digital business strategy, making it a high investment sector in the country. The popularity is growing to an extent of ‘AI washing’ — where the companies are applying AI labels to products and companies a little too generously.
Past trends have shown that although AI offers exciting possibilities, companies are more focused on building and marketing AI-based products rather than first identifying the need, potential uses and the business value they can generate. There is a tremendous increase in the number of startups and companies claiming to offer AI-based solutions, but are failing to execute. The underlying fault here could be largely attributed to the failure of adopting the right product mindset for AI.
Why Having A Product-Based Mindset Is The Key
A successful product is expected to have a consistent behaviour and to contribute to the over-the-top growth for a business. It is important to set and manage expectation of the users, gather their feedback and communicate these observations into new product offerings. However, while doing the drill for AI products, it may differ significantly from the traditional products.
For example, hardware or software products showcase a ‘deterministic’ behaviour, which means that a user’s behaviour is determined by product’s initial state and inputs, or is predetermined in most cases. However, in case of AI-driven products, it may not always have a deterministic behaviour, and may produce counter-intuitive results. This is due to the fact that a personalised recommender system may produce different results to a user action after learning additional preferences. It is therefore important for a product manager to have particular focus on strong product ideation and prototyping when it comes to AI products.
Also, hype around AI use cases are projecting more false-positive results than actual results. It therefore requires critical thinking to separate the hype from the real world. It is important to understand which products at the realm of AI can be commoditized and provide highest return on investment while overcoming the challenges.
“Using immature AI-based technologies and products is one of the several challenges to implementing AI”, Mosche Krank, CTO at Ness Digital Engineering had said in an interview with AIM. “Companies always overstate the actual capability, and it does not tell you those cases where the AI falls flat on its face.” He added that there is no substitute for experience. He also shared that your AI algorithm may work well on an engineer’s desktop but there is a need to productise the AI so that it can be deployed reliably in a system.
Having a clear product mindset also attract investors. They are on a lookout for product-driven companies with strong product-based mindset. Building a revenue generating product, bursting with innovative ideas and paying tiniest details to scale up the product are some of the characteristics that they are looking for in an AI-based company.
It can be said that having a product mindset for AI does therefore does three things:
- Fosters a culture that embraces disruption
- Brings out the creativity in organisation
- Eliminates barriers to innovation
Road Less Taken
It goes without saying that AI is a comparatively nascent field. Not all AI experiments may translate into field results. It may be challenging to work without any prior experience in the field and is important to have a thorough understanding of the current research to be able to build products.
In the overall process, getting the right amount data can be a hindrance. For AI products, access to data is the key. Once you have the data, it has to be cleaned, structured and labelled to train models. AI and machine learning concepts operate at a fundamental level and can lead to false positive it data fed is not robust.
Organisations must therefore look at their long term vision and embrace a holistic perspective to improve the field service industry with AI.
On A Concluding Note
With a surge in products like cloud, smartphones, IoT and others, the definition of products has changed quite significantly. Given the tech disruption, these trends will continue and they, in-turn will change the way products are made and used. It is therefore important to shift our thinking to view artificial intelligence as an evolution rather than as a revolution.
With a growing popularity in AI, it is important to build a mindset where AI professionals need the creativity to imagine how the technology can be applied, paired with the analytical acumen to measure results to determine success over time. They must be willing to take risks and perform experiments while being resilient enough to fail fast and move on faster.
Try deep learning using MATLAB