As AI has continued to develop throughout the year, and businesses have begun to understand more about the opportunities this technology has to offer, its influence will only continue to grow in 2025 and beyond. We asked experts across the technology industry how they think AI will continue to transform the wider technology and business landscapes in 2025 – here’s what they had to say.

Driving accessibility in AI and enhancing system performance – Roman Chernin, Co-Founder and CBO of Nebius

Looking ahead to 2025, the AI industry is set for rapid evolution, driven by both opportunity and challenges. We will see that access to compute resources will remain uneven, with challenges on both sides of the spectrum – access to super clusters for large enterprises and AI labs, and access to the newest chips at reasonable prices for research teams and individual developers and small startups. 

In this context, specialist providers like Nebius will have a bigger role to play in ensuring that everyone – not just those who can afford to spend big on cloud computing capacity – can access the possibilities of AI.

As AI continues to advance, we'll see more focus on developing specialised chips designed specifically for AI tasks. These chips will make AI systems faster, more efficient, and more affordable to deploy across various industries. Having hardware for specific AI functions will enable businesses to run generative AI solutions at scale without the high costs or performance issues that come with general-purpose hardware. This will enable a wider range of applications, from healthcare to entertainment, and could make advanced AI more accessible to companies of all sizes, and for their final customers.

The rise of AI-enabled cyber-attacks – Paul Holland, CEO of Beyond Encryption

Research revealed that 61% of organisations saw an increase in deepfake attacks over the past year – making it clear that the rapid rise of AI and deepfake technology will be a key issue for businesses in 2025.

AI has the power to create highly sophisticated deepfakes, which easily replicate voices, faces and identities. Fraudsters are increasingly using this technology to bypass traditional security systems by creating synthetic identities and using voice replication to trick individuals and organisations alike. This threat highlights the importance of safeguarding customer data and maintaining trust in all digital interactions.

Businesses must invest in advanced fraud detection systems to identify anomalies, and as deepfakes continue to become easier to create thanks to AI’s development – this should be a central part of security strategies. Alongside this, organisations must also implement robust multi-factor authentication to prevent unauthorised access. Finally, it’s vital to prioritise customer education to ensure that all individuals can spot a deepfake – allowing consumers to stop their data from falling into the wrong hands.

AI is also fuelling more sophisticated phishing scams, with 40% of phishing emails now crafted using this technology. This technology has opened a new realm of opportunities for threat actors – allowing them to carry out attacks with greater accuracy and a higher rate of success by leveraging Natural Language Processing to mimic writing styles, as well as automating stages of their attack chain.

As AI-powered cyberattacks become more prevalent, companies must reassess their cybersecurity measures to ensure their systems are robust and capable of addressing these evolving threats. Implementing advanced anti-phishing tools, multi-factor authentication, and secure email solutions is critical for safeguarding sensitive data against increasingly sophisticated attacks.

Agnostic AI – the driving force behind business success in 2025 – Kasia Borowska, MD and Co-Founder of Brainpool AI

This year, in a rush to capitalise on the AI hype, almost half of businesses leveraged off-the-shelf AI solutions. Although these tools offered quick wins for businesses with low upfront costs and straightforward implementation – they have prevented organisations from scaling AI and unlocking its true potential due to vendor lock-in.

Vendor lock-in means that businesses will have very limited flexibility. When integrating with a vendor, businesses must upload all of their data into the infrastructure of their chosen provider – which is a very time consuming process. This greatly limits a business’ flexibility as not only will they find it difficult to scale, they will also be subject to that vendor’s price increases and it will be difficult for them to switch providers due to compatibility and migration challenges.

For businesses to achieve success with AI in 2025, they must resist the urge to leap without looking. By adopting Agnostic AI, businesses can cut costs in the long term and respond more effectively to industry and regulatory changes. Most importantly, it will enable businesses to develop solutions that will unlock true value as Agnostic AI allows organisations to mould each use case around their unique business needs to secure long term success.

It is clear that AI’s evolution will bring both opportunities and challenges in 2025. From expanding access to computing resources to adopting agnostic AI – businesses must act strategically to reap the full benefits. With careful planning, 2025 will be a year that truly redefines how AI shapes and transforms businesses.

For more startup news, check out the other articles on the website, and subscribe to the magazine for free. Listen to The Cereal Entrepreneur podcast for more interviews with entrepreneurs and big-hitters in the startup ecosystem.