All the problems discussed in the previous posts can create two major cross-sectional problems: the likely event to run out of money before hitting relevant milestones toward the next investment, as well as whether pursuing specific business applications to break even instead of focusing on product development.
In terms instead of classifying different companies operating in the space, there might be several different ways to think around machine intelligence startups (e.g., the classification proposed by Bloomberg Beta investor Shivon Zilis in 2015 is very accurate and useful for this purpose). I believe though that a too narrow framework might be counterproductive given the flexibility of the sector and the facility of transitioning from one group to another, and so I preferred to create a four-major-clusters categorization:
i) Academic spin-offs: these are the more long-term research-oriented companies, which tackle problems hard to break. The teams are usually really experienced, and they are the real innovators who make breakthroughs that advance the field;
ii) Data-as-a-service (DaaS): in this group are included companies which collect specific huge datasets, or create new data sources connecting unrelated silos;
iii) Model-as-a-service (MaaS): this seems to be the most widespread class of companies, and it is made of those firms that are commoditizing their models as a stream of revenues. They can appear in three different forms:
1. Narrow AI — a company that focus on solving a specific problem through new data, innovative algorithms, or better interfaces;
2. Value extractor — a company that uses its models to extract value and insights from data. The solutions usually provided might either integrate with the clients’ stack (through APIs or building specifically on top of customers’ platform) or otherwise full-stacks solutions. All the models offered can be trained (operative models) or to be trained (raw models);
3. Enablers — a company that is enabling the final user to do either her own analysis (all-in-one platforms), or allowing companies to make daily workflows more efficient, or eventually unlocking new opportunities through the creation of intermediate products (e.g., applications).
iv) Robot-as-a-service (RaaS): this class is made by virtual and physical agents that people can interact with. Virtual agents and chatbots cover the low-cost side of the group, while physical world systems (e.g., self-driving cars, sensors, etc.), drones, and actual robots are the capital and talent-intensive side of the coin.
The results of this categorization can be summarized into the following matrix, plotting the groups with respect to short-term monetization (STM) and business defensibility.
Starting from the more viable products, the MaaS are the companies with the highest potential to monetize their products in the short term, but also the less defensible. DaaS on the other side is way less replicable, and highly profitable anyway. Academic spin-offs are the long bet, which is based on solid scientific research that makes them unique but not valuable from day one. Finally, RaaS companies are the ones who might face more problems, because of high obsolescence in hardware components and difficulties in creating the right interactive interfaces. This classification is not intended to rank any business based on how good they are, and it does not imply that specific companies belonging to specific classes are not going to be profitable or successful (e.g., X.ai is a high profitable company with a great product into the RaaS area). It is nothing more than a generalization tool useful to look at the sector through the correct lenses.