Scientific progress is rarely linear. Only sometimes do frustrated lab notes and hours reviewing literature give way to a discovery. Big ideas burst into view unexpectedly, the freshly published, prestigious journal article rapidly collecting an ever-growing impact factor. Some topics go further, bolstered by wider media coverage and glowing headlines, and researchers are invited to speak not just at conferences but on mainstream TV shows. Then, as interest heightens and funding flows, promises stretch further than the results – cooling sets in. New results are sparse and theories are disproved. The initial research may stand, but it simply does not live up to the hype.

“The poster child of such over-promise is string theory”

The poster child of such over-promise is string theory. Introduced in the late 20th century as the “Theory of Everything,” string theory’s mathematical beauty and potential to unify general relativity and quantum mechanics saw it trickle from the offices of mathematicians into BBC articles, mini documentaries, and Rick and Morty. But despite decades of development, it remains untested and, many argue, untestable. It has enriched mathematical physics, no doubt, but it stands as a cautionary tale of over-investment into theories that people desperately want to work.

The current physics breakthrough dominating the headlines is one of a different nature. Quantum computing, a new kind of computing that uses the principles of quantum mechanics to store and process information in ‘qubits’ rather than binary ‘bits’, is more experimental. Its goal is a transformative leap in computation, since qubits can exist in a superposition of many states (instead of just ‘1’ or ‘0’), and this means certain tasks might be carried out in a much faster fashion. Crucially, this isn’t a technological miracle – quantum computers are actually slower than regular ones for many processes, and consequences of their quantum nature (like entanglement or ‘linking together’ between different qubits) make them tricky to scale up, and incredibly delicate. Nevertheless, their hypothesised use-cases range from simulating atomic behaviour to breaking classical encryption systems.

“We just don’t have quantum computers big enough to do anything useful”

Recently, a rush of investment with enormous funding rounds and public-facing buzz words has ushered it in as the technology the world is watching. However, despite scientists’ best efforts, the reality is more tempered. Qubits remain fragile. Error correction is a monumental challenge, since qubits can’t simply be reset to a classical ‘1’ or ‘0’ digit when affected by noise. Scalable architectures are still experimental – we just don’t have quantum computers big enough to do anything useful!

Less than a year ago, in May 2024, the communications company Zoom posted that it was “now offering post-quantum end-to-end encryption in Zoom Workplace”. On the same page, it admitted that “quantum computers that are powerful enough to run these [decryption] algorithms are not known to exist”. This level of caution for an as-yet untested field is incredible for a videoconferencing service, and as for what exactly ‘post-quantum’ means, your guess is as good as mine. I take it as Zoom pre-empting a hypothetical future where quantum computers are scaled up, classical encryption is broken, and my physics supervision zoom-calls are released – finally exposing my questionable integration skills. Except, in their own cryptography documentation, Zoom admits that they “do not currently defend against … quantum computers powerful enough to break classical cryptography”. Oh…

“A follow up paper from Microsoft with more proof is mysteriously non-existent”

Even at Microsoft, a recent discovery highlights the uncertain line between genuine progress and company incentives for marketing their discoveries. In February 2025, Microsoft announced Majorana 1, a quantum computer chip far more resistant to information loss. Microsoft stocks soared, but a few months later, criticism is still rife. Physicist Dr Henry Legg raised concerns about result reproduction and protocol flaws, and a follow up paper from Microsoft with more proof is mysteriously non-existent.

Does this mean quantum computing might just be string theory 2.0? Well… no. For one, the very process of building quantum computers has already driven advances in condensed matter physics and non-equilibrium thermodynamics. So will, as one optimistic Judge Business School speaker recently proposed at a Cavendish conference, everyone soon have a quantum computer in their home office? Also no. This leaves us with a more structural question: how should we fund and support science in the age of the hype cycle?

Hype draws money and talent into areas that genuinely need both, but it can also distort expectations, pressure scientists to overpromise, and sideline equally essential research. Together, this fuels a growing rhetoric of anti-intellectualism. When scientists competing for limited funding are systematically encouraged to make bold claims about project outcomes, public attention is more likely to judge the value of their work by these ambitious, utilitarian goals, and to criticise the results if they fall short.

But let me repeat: scientific progress is not linear. You likely wouldn’t be reading this article if some particle physicists had not decided to streamline their information-sharing systems and so invented the world wide web. Research does not need to meet a predefined project outcome to be worthwhile. The challenge of moderating hype cycles is not eliminating excitement – it is maintaining rigorous scientific review without fuelling public disillusionment. Done right, this advances human understanding – and creates some genuinely cool stuff along the way!