Big brands are investing more and more in Artificial Intelligence (AI), considering its transformative role for the future of work. But what really changes with AI in companies? Find out about part of this world with Ricardo Parreira, CEO of PHC Software.
PHC Software is a Portuguese multinational dedicated to the development of management software. Founded in 1989, it has over 200 employees and more than 400 certified partners. It has more than 35,000 clients in 25 countries and more than 150,000 software users. PHC has offices in Lisbon and Porto (Portugal), Madrid (Spain), Lima (Peru), Luanda (Angola) and Maputo. The company entered Mozambique in 2000 and has been growing in this market: in the last year, sales have increased by 40 per cent, the team has doubled and the partner network has expanded to cover the whole country.
PHC Software recently presented a virtual assistant called “Cris”. This is a type of generative AI for business management, which can assist in the daily workflow, both in the areas of compiling and analysing data and in filling in forms. The tool is also capable of creating automatic solutions, tailored to each team, and opens up a world of increased efficiency for managers, adapted to the reality of each organisation. The CEO of PHC Software, Ricardo Parreira, talks about this technological added value, taking a look at the impacts that the evolution of AI solutions will have on the transformation of companies, people and the economy.
What are the different uses of generative AI? What are the specific applications and relevance for users, both companies and individuals?
Generative AI is perhaps one of the most impactful technologies for companies ever, because a large part of the tasks performed based on people’s knowledge can be greatly expanded. That’s what we believe: people can do other things, they can dedicate themselves to what they do very well. We presented two studies, one by Harvard Business School and the other by MIT [Massachusetts Institute of Technology], showing that, in certain tasks, we can be up to 37 per cent faster and 40 per cent better at performing tasks.
Specifically in Mozambique, how can this technology be capitalised on, given the characteristics of the companies and the country?
Employability is one of the great ghosts that arise when talking about AI. I think this has always been the case with disruptive technologies. When computers came along, it was also said at the time (1960s and 1970s) that lots of people were going to lose their jobs because they were just “stamping paper” all day. However, computers created many more jobs. When the Industrial Revolution came along, it was also said that jobs would be lost, but in the end, many more were created. I believe that with AI, many more jobs will be created and the current ones will be better performed. I often say that AI isn’t going to destroy jobs. I’m sure that if someone loses their job, it will be to someone who has mastered AI. I hope that there is strong competition in the market and that it doesn’t remain in the hands of a handful of companies, as is the case with GPT Chat [OpenAI’s AI tool] at the moment.
But AI also has inaccuracies that have led companies to be extra careful. I’m referring, for example, to the likelihood of producing wrong information or answers, which can lead organisations to undesirable decisions and results. What is the real degree of risk?
AI is still in its infancy and engines sometimes have “hallucinations”. In other words, they suddenly give totally wrong information, but with all the arguments well written. So we have to be very careful not to assume that it’s a platform that holds the absolute truth, but that it’s just a tool that can be useful in our activities as human beings. Another very interesting topic when talking about AI is related to sources. Many of today’s systems are already working to reveal search sources immediately, i.e. for the system to give three or four answers and tell you the sources. These systems have learnt from Internet content that, in most cases, was created in the Western world. So the bias is very strong. On the other hand, it’s called GPT [Generative Pre-training Transformer] because it has been ‘pre-trained’, i.e. humans have indicated which answers are the right ones. If those humans were biased, it’s inevitable that the AI will be too. For this reason, I use AI very little for political issues, but from a management point of view, I haven’t picked up any major biases yet.
“It’s not AI that’s going to destroy jobs. If someone loses their job, it will be to someone else who has mastered AI. I expect there to be strong competition in the market,” Ricardo Pereira
This tool is also very useful for the world of accounting, which I believe is going to evolve radically over the next few years. In our software, if you ask for an income statement for the last three years, ChatGPT will tell you where things are going well, how the working capital is, for example, how the figures are evolving and why.

How has PHC dealt with data protection when accessing AI?
We use a security system developed by Microsoft, because this brand realised that the issue remains the number one obstacle to the adoption of AI. So they have prepared all the technology to protect the data. In our case, as PHC, we don’t communicate directly with ChatGPT, we go through Microsoft first, which works on the data so that the AI doesn’t use the data for training and doesn’t keep it with a source ID. I often say that this is the technology with the biggest marketing campaign in history, but it seems to me that we talk more about the problems of privacy, security or the fact that this could end the world, than we do about the benefits and spectacular examples of what can be done. However, I recognise that regulation is essential, inevitable and urgent. Governments cannot abdicate their role in this. We didn’t vote for technology companies to define our future, we voted for governments. OpenAI [the owner of ChtGPT] has already raised the urgency of regulation with the US Congress. The European Union is already very close to having it, China was the first to regulate, but the United States is still a little way off, although very concerned.
What does PHC offer companies and other users of its AI systems?
Developing a ChatGPT is something that is only within the reach of a company with a huge level of funding, OpenAI, which developed the tool and has received more than a billion dollars in funding from Microsoft alone. OpenAI now has two million companies using their software and we are one of them. We use Microsoft’s data security system, called Azure AI, to protect the confidentiality of company data, and we have embedded ChatGPT within our software because, to be really effective, AI needs information. The combination of management software and AI is explosive.
“Mozambique is well positioned as a power to provide a transitional energy source, gas. I maintain the view that the country is in a brilliant position to do well”
We have embedded generative AI within our software, where there is data, and the user can ask for help, can ask questions, can ask to perform various tasks, among other functionalities. We launched our system here in Mozambique on 28 September 2023 and it is available to our thousands of customers. We have an integrated system, within our software, to work wonders. We now also have human capital management software, where we can put performance appraisals. Each person can receive their evaluation, find out what their strengths and weaknesses are. You can find out how to deal with weaknesses. Among the various functionalities, I can, for example, tap a button and ask you to send me an e-mail to charge a certain customer. Let’s say there’s no more patience for that customer, that it’s really hard for them to pay. Then we can change the tone. I just need to tell the software to send a message in a more serious tone. The AI writes everything down in five seconds, without being angry or rude.
What is the possibility of success in this area in Mozambique?
The first feeling I had in Mozambique was one of immense curiosity and enthusiasm, although I was expecting to feel more fear. In other countries I felt more fear than enthusiasm. The curiosity is also very much related to the fact that people have a feeling that it really does serve a purpose, but they still lack practical cases. And so, with this enthusiasm in Mozambique, the value of AI systems can be better realised. People and companies want to make better use of what can be done. In order for customers to make better use of these tools, we provide training and certification and are training partners so that they too are prepared to train customers, because this is a key point. We only sell through partners. We can modify the software, configuring it to the customer’s specific needs. All this use of generative AI is within this context that they can use. People can automate the most different tasks.
This is a disruptive area and I don’t think there’s much competition yet. But it’s coming up: how does this change the scenario for you?
I think this moment is going to be a radical change for our business. It was obvious that technology is a differentiator, but this reality is going to be explosive. In a year or two, companies that don’t use generative AI are going to be left behind because, in the theory of innovation, there are always early adopters. I think this innovation curve is going to be very accelerated – and we’re still using it in a very basic way. I think there are some very interesting things coming. One of them is the concept of agents. There are already entire companies and some CEOs who are AI. So I think there are going to be a lot of opportunities from that point of view.
By: Pedro Cativelos – Photo: Mariano Silva & IstockPhoto