How to use AI in your startup when you have limited resources

Circuit board and AI micro processor, Artificial intelligence of digital human. 3d render
Circuit board and AI micro processor, Artificial intelligence of digital human. 3d render
Circuit board and AI micro processor, Artificial intelligence of digital human. 3d render

Mark my words: the first one-person unicorn company is launching soon. This unprecedented feat will only be made possible by the transformative power of AI.

For startups and small businesses, every dollar and every hour counts. You have limited resources compared to larger competitors, making any productivity or efficiency gain all the more significant.

This is why AI shouldn’t just be a buzzword. When adopted, it can be transformational for smaller businesses, and some of those may go on to be worth billions.

Generative AI can democratise access to knowledge, skills, or opportunities that might otherwise be beyond the reach of SMEs with limited resources and influence.

For example, our new AI-powered, self-serve platform for founders has opened up opportunities to kickstart a capital-raising conversation like never before.

Using AI tools and real-time data, it streamlines the capital-raising process for both founders and investors. It has reduced a lengthy, manual process that previously took weeks to just a few hours, redefining the digital capital-raising journey.

AI also democratises innovation.

As Forbes reports, AI can help foster divergent thinking, assist in refining ideas, and help resolve complex problems.

Beyond a productivity boost, AI is making a way for time, our rarest commodity, to be used more efficiently.

The 1-way v 2-way door choice

It’s clear that AI is unstoppable, and as Elon Musk pointed out at Cannes Lion 2024, “only the companies that effectively implement AI will survive.”

Thus, many leaders are now at a crossroads: How can I implement AI in my startup? Should we go out-of-the-box or bespoke?

Startup land is fast-paced and often requires founders to make decisions quickly.

A useful way to address these choices is Amazon founder Jeff Bezos’s approach to categorising decisions: “One-way doors” are almost impossible to reverse, such as choosing to pivot your core business model, and “two-way doors” are easy to reverse, and can include decisions around experimenting with generative AI to create marketing content or integrating out-of-the-box AI tools to enhance customer support.

In a recent Telstra/MIT study, while only 9% of companies had adopted GenAI widely in 2023, 76% had already been trialling and working with it in some way. Business leaders should seize the opportunity to experiment with two-way door decisions such as leveraging third-party and out-of-the-box AI solutions like Google’s Gemini or OpenAI’s ChatGPT, which are more affordable (time and resource-wise) compared to building bespoke AI functionalities from scratch.

It’s a low-effort and low-consequence decision, where an application for most internal company functions can be found and implemented quickly and painlessly.

If you’ve got some technical team members, you can explore relatively easy-to-implement solutions by accessing powerful open-source models on AI community platforms like Hugging Face. Although it may seem daunting at first, implementing these solutions is becoming easier than ever.

Conversely, one-way door decisions require a significant amount of time, effort, and money. Many companies are currently implementing ‘two-way door’ solutions and often incorrectly branding themselves as AI companies.

To truly be considered an AI company, rather than just a company that uses AI, your core business model should be delivering disruptive and proprietary AI solutions. This requires deeply embedding AI throughout your entire infrastructure—a one-way door decision that commits you to this path for the long term.

Here’s what I’ve learnt throughout VentureCrowd’s process of building our own AI capabilities without losing sight of our core value proposition.

How to kickstart the AI adoption process  

1. Start with good data

Data is the foundation of your AI infrastructure. To adopt AI, you must have sufficient high-quality data and implement proper data governance from the start.

Robust documentation and a data-driven culture, often overlooked but critical, are vital for ensuring transparency, accuracy, and reliability in AI.

It enables effective training, validation, and troubleshooting of AI models. If you don’t have the right documentation in your processes, it’s difficult to apply useful AI solutions to your business.

2Have a clear purpose

Ensure you have crystal-clear clarity on why you are seeking to implement AI in the first place. What is the high-impact, high-value problem you are trying to solve?

Then, ensure that everyone understands what problem you’re trying to solve at a business level. If the real problem you are trying to solve via AI is not clearly laid out in your business and technology roadmap, it is going to be difficult to implement, and it increases the chances of the implementation becoming an ad hoc project that could die or not be effective without proper alignment.

Ask yourself: Are you seeking to automate existing processes or augment your team’s capacity via AI? What are the key pain points that you need to tackle as a business? How can technology help you and your teams get there? What does success look like? Aligning digital products with commercial problems will drive ROI.

Technology investment must align with a clear strategy and use cases. If alignment from the top is missing, there is no AI or technology solution that will help.

3. Evolve your culture

Beyond upgrading your technology, you also need to upgrade the team’s mindset. This means establishing a culture of innovation and embracing the unknown because the AI adoption journey will present you with a lot of unknowns.

According to McKinsey, companies with strong innovation cultures are much more likely to succeed in digital transformations. They’re also ahead of their peers in using technology to distance themselves from competitors.

While startups already have innovation in their DNA, they should be intentional in maintaining this mindset and establish agile rituals when embarking on the AI adoption journey.

Ensure that you’re developing the right talent and skill set and encouraging people to experiment. Allocating time for your team to continuously upskill is not a nice-to-have if you want to stay ahead of the game. AI developments are fast-paced and the arena is changing every week.

Finally, ensure your AI initiatives are spearheaded by a designated AI champion and that they have the backing of the executive team. This role requires a mix of technology and communication skills to help executive teams understand the benefits of experimentation.

AI is not going anywhere. All the research indicates higher adoption of GenAI over the next few years as well as AI becoming increasingly more powerful and useful. We’ve only begun to scratch the surface.

It’s important for small businesses and startups to make a head start by learning and experimenting. Fix your data, be clear about your purpose and evolve your culture to embrace the potential that AI offers.

Most importantly: Just start. More often than not you’ll realise it’s a two-way door.

By Diego Mogollon, CTO of crowdfunding platform VentureCrowd

This article was first published by StartupDaily