SUGGESTIONS on the government’s intentions to expand the use of artificial intelligence (AI) in disaster management systems are captivating.

But they could as well be somewhat disheartening, as ability to see disaster coming up or predicting its scale once it starts unfolding would equally require an equally advanced response mechanism.

That is where problems crop up as to the use of AI tools to predict disaster or map out its scale and then revert to a ‘business as usual’ stance when setting out to respond to what is happening.

The method would be worth deploying if it creates a disaster preparedness guarantee chain reaction, where the ‘sniffing’ of tragedy leads to the outlining of needs without let or hindrance.

What is needed is a prior design of what is affordable, meaning that an AI design rapidly indicates what is needed on the spot – from blankets, tents, water, food, etc.

The wish to roll out AI tools in that regard was aired in Zimbabwe at a recent preparatory meeting of ministers responsible for disaster risk management in Southern African Development Community (SADC) member states. 

There is no doubt that the tools would improve early warning mechanisms and strengthen response to increasingly complex natural other emergencies, but the issue is whether this ability is sufficiently backed by a reaction mechanism.

Outside that aspect it becomes a white elephant – a typical case of having sharp tools to tell the extent of looming flooding and then asking smallholder farmers to fend for themselves.

What is likely to hinder adding the supply chain component to sharp prediction ability in relation to disasters is an innate disposition at most levels of government to do things manually and by whim once there is need for some significant disbursement.

Nearly all such disbursements could have portions of procurement being inflated in pricing and thus facilitating part of the cash to go around, implying that bureaucrats would wish for tools to make precise predictions while retaining the reaction process in their hands.

That would mean tying matters to transparent, digitally determined necessary procurement of materials needed for urgent and effective disaster relief plan of action.

There could even be times of people and institutions sympathetic with affected families being asked to direct assistance funds or materials to the relevant local authorities at the regional or district level instead of assisting the victims directly.

It is hard to understand how those issuing such directives would fail to appreciate the fact that people can read between the lines and map out exactly why assistance needs to be directed at bureaucrats rather than the victims themselves.

At least supposedly, only through the latter option could such assistance be equitably distributed.

Disaster management executives thus ought to understand that equipment of a specific kind work chiefly with the relevant forms of management and administration of resources.

It is commonly said that the handloom for cloth fabrication gives one a feudal system (of owners and labourer at the loom), while the steam mill demands capitalist relations where equipment and labour are procured and land is mortgaged.

Digital prediction needs fully transparent procurement, pricing of materials and urgent delivery as otherwise it is of little or no use.