1.🚁 Heli view: How AI opens up new battlefronts

This early in the technical evolution it’s rare to find game-changing examples of AI in Africa’s green economy. But there are some undeniable ones. 

At the frontier: Mini-grid operator Husk Power Systems uses predictive AI in Nigeria to forecast supply and demand. 

• The company plans to expand solar power to 7.7 million Africans in five years, using 1,500 mini-grids.

• AI reduces staffing needs by 40% to a team of 2,500 operators. 

• It also reduces the use for back-up diesel generators by 25%. 

• And utilisation of solar assets is increased by 10%. 

The promise: AI acts as a catalyst, optimising resource use and reducing waste.

• Estimates suggest a global GDP boost of $16 trillion in 2030, about 10%.

Local angle: Transformative applications should also benefit Africa's green economy.

• Andy Kuper, the CEO of LeapFrog Investments, a major green funder in Africa, recently said: “AI has immense potential to power decarbonisation.” 

• He named “everything from agriculture to energy grids to carbon tracking. In all these use cases, smart data-sharing and analytics can drive costs down for emerging consumers and producers in their millions – if used responsibly.”

The dangers: No new technology comes without its own set of risks. 

• Experts warn of job losses and human enslavement by machines. 

The problem: Justified debates about AI threats obscure another, less sexy debate – how hard we’ll have to work to get machines to do our work.

• It's becoming clear AI doesn't mean putting your feet up on a desk. 

• AI gains are worthwhile but laborious (see our cheat sheet below).

How: The point is made by AI pioneer Nigel Toon in his new book “How AI Thinks”.

• He writes, "AI is the most powerful tool we have ever created." 

• "Those who take the time to understand how AI thinks will end up inheriting the earth." The emphasis is on the "how".

Differing opportunities: AI is a platform technology. Applications are specific to sectors – including green ones. Everyone has to figure out their own. Prime candidates are: 

• Renewable energy

• Carbon & biodiversity

• Sustainable agriculture

Renewable energy: AI can help to optimise power grids, forecast demand and integrate renewable sources effectively.

• “Predictive maintenance” also flags repairs before major problems arise while avoiding unnecessary replacement of parts that are still in good condition.

• A Nigerian study says “Automated systems adjust energy distribution and dispatch resources efficiently, preventing grid instability and optimising overall energy use.” 

Carbon & biodiversity: African ecosystems face threats from habitat loss to poaching. AI provides tools for more effective conservation. 

• Utilising drones, acoustic monitoring and satellite imagery can boost the effectiveness of interventions to protect species and landscapes.

• Hluhluwe–Imfolozi in South Africa has become a “smart park” processing inputs from cameras to alert rangers to potential rhino poaching.

Sustainable agriculture: Farmers can use AI to optimise existing practices and increase yields substantially.

• A Nigerian study talks of, "Using AI to precisely monitor crops, evaluate soil health and improve decision making through weather forecasting." 

• Another focus is better resource allocation for water and fertilisers by leveraging real-time data insights.

• Machine learning algorithms can help time local planting optimally. 

Healthy scepticism: Is using such technology out of reach for smallholders involved in subsistence farming?

• Larger landholders will benefit predominantly. Where they can roll up less efficient, smaller farms, employment will suffer, creating lower median incomes.

• However, in Kenya, AgriTech Analytics helps smallholders by using AI to detect pests and advise on precise countermeasures.

2. Cheat sheet: How to build your own climate AI today

Everyone has played with ChatGPT, but have you used it for anything other than mundane tasks?