- Article
![](/-/media/media/global/images/articles/why-reliable-data-can-be-an-opportunity-for-businesses-banner.jpg?h=693&iar=0&w=1440&hash=5F08395D02C61922DA79EC26ADC4CA77)
- Innovation & Transformation
- Digital transformation
Why reliable data can be an opportunity for businesses
Artificial Intelligence (AI) and machine learning software have the potential to transform productivity in the workplace, boosting margins and improving efficiency across multiple industries. Yet their rise also poses a challenge. As businesses across Asia become more dependent on digital technology, they also become more dependent on data.
In the age of deepfake technology and advanced cybersecurity attacks, protecting the integrity of that data will be a priority for all businesses, especially those operating across borders.
Understanding the data challenge
Each day, over 300 million terabytes of digital information are created around the world. Data generation is growing at over 20% a year.¹
As the volume of data increases, so does the challenge of managing it. While accurate, timely and reliable data can hold valuable insights, error-strewn or misleading data is potentially costly. To prevent this, businesses should focus on maintaining data sets that are fit for purpose and relevant to their operations, rather than spending time and resources on data that will never be used. Additionally, it is important to strike a balance in data retention, ensuring that necessary information is preserved without over-retaining data, which can lead to increased storage costs and potential compliance risks.
Access to trusted data is especially important for the growing number of businesses looking to deploy AI and machine learning applications. AI models can process far more data than is humanly possible, but the quality of their output depends on the information being used. Data scientists often sum it up as: garbage in, garbage out.²
Sifting the good data from the bad may not be as straightforward as it seems. Businesses need to spend time cleaning and sorting through their various data sources to remove errors that could mislead an AI system. Incomplete or outdated data sets, or too much irrelevant information, risk producing confusing results.
Decision-makers also need to be aware of potential biases in the data. A common mistake is to train an AI model to validate an existing theory, which can lead to useful data being ignored.
Businesses that are looking to analyse their own proprietary data sets – such as sales numbers or customer searches on their websites – need to be aware that even data from the most trusted sources can become misleading if it is not handled correctly. A sales model trained during the height of the pandemic, for example, would not be applicable today.
Off-the-shelf AI tools may be a cost-effective option for smaller businesses. Where businesses deploy these solutions, however, they need to ensure that the right sources have gone into their development.
Similarly, businesses that rely on third-party data, such as from their suppliers, can consider additional steps to review its accuracy and trustworthiness. This could be as simple as asking suppliers for additional information about their data collection processes, or seeking independent verification – for instance on supply chain emissions.
Safekeeping and compliance
The AI era also opens new opportunities for malicious actors. New tools allow fraudsters to create convincing images, audio, or video files that can be used to dupe businesses into sending money to fake suppliers or spread disinformation to cause reputational damage. The growing dependence on data also makes it more valuable for hackers and cybercriminals, leading to growing numbers of ransomware attacks.³
This poses new challenges for businesses that handle sensitive data and rely on the trust of their customers to operate. In one McKinsey survey, nearly 10% of businesses said they had stopped dealing with a supplier after learning of a data breach.⁴
Cybersecurity protocols and training to mitigate against these threats – from external or internal parties – will be essential in maintaining the integrity of AI models and data sets.
Businesses can also consider incident response plans and regular risk assessments to detect threats. This may involve working with cybersecurity experts to identify vulnerabilities and ensure third-party risks are mitigated. For ongoing protection, outsourced vendor solutions can be cost-effective, particularly when considered against the potential cost of a cyber-attack. Global spending on cybersecurity vendors is set to rise by 13% a year up to 2025.⁵
Businesses also need to be aware of the growing regulatory scrutiny of data sources and data security since the European Union’s General Data Protection Regulation (GDPR) took effect in 2018. In the past two years, several countries in Asia Pacific have introduced or updated their own data privacy rules, including Australia, China, India, Indonesia, Japan, Thailand, and Vietnam.⁶ India is the latest major jurisdiction to do so with the Digital Personal Data Protection Act in August 2023.
Notably, many of these recent moves restrict the transfer of data across borders. That poses a particular challenge for businesses with operations across Asia Pacific who need to ensure that they are compliant with the latest regulations when looking to consolidate and analyse country-level data.
How trusted data helps businesses
Gathering timely, reliable, and secure data could opens a range of potential benefits to businesses in all industries:
- Trusted data is critical to unlocking the potential of AI. This allows companies to harvest further gains from greater efficiency, improve cash flow management, and refine their analysis of customer needs.
- Investing in sound data protocols lowers compliance and reputational risks, leading to better credit profiles and improved access to funding. One example from HSBC is our partnership with cross-border e-commerce data platform Dowsure Technologies to provide data-led trade finance for Amazon merchants in mainland China and Hong Kong.⁷
- Implementing internal security systems and educating employees in building trust can help upskill and improve retention through involving every layer of the workforce in the effort.
- By investing time and effort in building trusted data networks, businesses can deepen and strengthen ties with trade, supply chain, and customer networks that will reap dividends for many years to come.
Preserving trust in the future
New threats need new solutions. The development of quantum computing, for example, could render existing encryption and security safeguards redundant, leaving businesses exposed to increasingly sophisticated cyberattacks. To prepare for this eventuality, HSBC is already deploying quantum key distribution in our own business, and testing quantum protection for trading processes.⁸
Banks like us, and other financial services firms, depend on trust to operate – and have a responsibility to protect our customers, our staff, and our investors.
But there are also exciting opportunities to use advanced technology to enhance trust and security in data processes. We already use blockchain and distributed ledger technology, for example, to verify digital trade documentation, speeding up cross-border trade financing.⁹
Given the rising importance of trusted data in key decisions, all businesses need to consider steps to maintain the integrity of their data now, to protect over the long term.