A complex scientific problem that took microbiologists a decade to unravel has been cracked in just 48 hours by an advanced artificial intelligence (AI) system developed by Google.

A Decade of Research Solved in 48 Hours

In what many are calling a revolutionary moment for science, researchers at Imperial College London were somewhat stunned when Google’s AI tool, aptly named ‘co-scientist’, managed to solve a mystery that had challenged microbiologists for ten years. The team, led by Professor José R. Penadés, had dedicated years to investigating how superbugs (bacteria resistant to multiple antibiotics) developed their dangerous immunity.

Tails

The crux of their research focused on understanding how some bacteria could acquire ‘tails’ from viruses, enabling them to transfer resistance between different species. This process is akin to bacteria acquiring ‘keys’ that allow them to move between hosts, posing a severe risk to global health.

However, when Prof Penadés submitted a simple prompt to the AI system, without feeding it unpublished data, the tool not only replicated the team’s hypothesis but did so in less than two days.

Four Extra Hypotheses

Incredibly, the AI went further than simply replicating the team’s conclusions and generated four additional hypotheses, all of which, according to the researchers, were scientifically plausible. One of these entirely new insights is now actively being explored by the team, potentially opening up uncharted avenues in the fight against antibiotic resistance.

How Did the AI Crack the Code / Confirm Their Hypothesis?

The AI tool behind this breakthrough, developed by Google DeepMind, was designed as a collaborative assistant rather than a full replacement for human researchers. Branded as a “co-scientist”, the system is purpose-built to aid scientists in hypothesis generation, experimental design, and data analysis.

Rather than simply trawling publicly available data, the AI can synthesise information from a range of inputs, including academic papers, scientific databases, specialised AI feedback loops, and manually submitted private documents.

AI Can Navigate Through Scientific ‘Dead Ends’

According to Dr Tiago Dias da Costa, who co-led the experimental validation work, the true power of the AI lies in its ability to navigate through scientific “dead ends”. These are common in research, with scientists often spending months or even years testing hypotheses that ultimately yield no fruitful results. As Dr Costa points out: “AI has the potential to synthesise all the available evidence and direct us to the most important questions and experimental designs.”

The AI’s ability to eliminate unlikely paths and highlight the most promising ones could dramatically shorten research timelines, potentially bringing life-saving treatments to market much faster than current processes allow.

What Makes This Breakthrough Special?

Perhaps the most astonishing aspect of the discovery is that the AI system managed to reach a complex scientific conclusion without prior access to unpublished research. Prof Penadés initially suspected foul play, jokingly emailing Google to ask if it had somehow accessed his computer. The company confirmed that the AI had only used publicly available information.

This suggests that the AI was able to draw novel conclusions independently, which is something even seasoned scientists can struggle with, especially in fields as intricate as microbiology.

Supporting Scientific Discovery

Professor Mary Ryan, Vice Provost for Research and Enterprise at Imperial, has highlighted the broader implications of this breakthrough, saying: “The world is facing multiple complex challenges—from pandemics to environmental sustainability and food security. To address these urgent needs means accelerating traditional R&D processes, and AI will increasingly support scientific discovery and pioneering developments.”

What Are the Wider Implications?

The research team believes that if they had access to such AI capabilities from the outset, it could have saved them years of work. This has sparked a broader conversation about the role of AI in research in, for example:

– Accelerating discoveries. AI can help researchers rapidly test and refine hypotheses, cutting down on lengthy trial-and-error processes.

– Reducing costs. Speeding up research timelines could dramatically cut the financial costs associated with long-term scientific projects.

– Democratising research. AI could also help level the playing field, giving smaller research teams access to powerful analytical tools once reserved for larger institutions.

Concerns

However, the rise of AI in science isn’t without controversy. There are concerns over the potential loss of jobs and the diminishing role of human intuition in scientific discovery. That said, Prof Penadés offers a different perspective, saying: “It’s not about replacing scientists. It’s about having an extremely powerful tool to help us work smarter and faster. This will change science, definitely.”

A Glimpse into the Future of Scientific Research?

The implications of this breakthrough extend beyond the immediate challenge of antibiotic resistance. As the technology matures, AI systems like Google’s co-scientist could actually redefine how research is conducted across multiple fields, from climate science to drug discovery.

Google researchers suggest that AI could be used to accelerate the literature review process, one of the most time-consuming aspects of scientific research. By quickly analysing vast amounts of information, AI could help scientists identify gaps in existing knowledge and generate novel hypotheses at a rate previously unimaginable.

Also, partnerships like the one between Imperial College London and Google could become a model for future collaborations between academia and the tech industry. The Fleming Initiative, which focuses on combating antimicrobial resistance, aims to expand this model to other pressing global challenges, including:

– Developing rapid diagnostic tools for early detection of infections.

– Leading drug discovery efforts using AI-driven analysis.

– Building international networks of research experts to tackle global health crises.

Cautious Steps

While the technology is still in its early stages, this breakthrough has shown what’s possible when human expertise and AI capabilities work together. For now, researchers remain cautiously optimistic about what’s to come. As Prof Penadés put it: “It’s like playing a Champions League match with the best tools possible—we’re finally competing at the highest level, and the possibilities are spectacular.”

What Does This Mean For Your Business?

This apparently remarkable breakthrough, where Google’s AI ‘co-scientist’ managed to solve a decade-old scientific mystery in just two days, could signal more than just a milestone for microbiology, it could offer a glimpse into the future of scientific discovery and technological collaboration. By demonstrating the capacity to generate not only accurate hypotheses but also entirely new, scientifically plausible insights, AI has proven itself, in this case, as an invaluable asset in pushing the boundaries of human knowledge.

For researchers, the ability to bypass years of trial-and-error, sidestep scientific dead ends, and fast-track promising avenues of investigation could redefine research timelines across countless fields. For example, no longer will progress be bound by human limitations in data processing and analysis. Instead, AI will enable researchers to focus their expertise on refining experiments and validating results with unprecedented efficiency.

The significance of this breakthrough may also stretch far beyond the scientific realm. For businesses, particularly those looking to harness AI to drive growth and innovation, this development offers a lesson in that AI’s greatest strength lies not in replacing human insight but in amplifying it. Companies hoping to leverage AI for commercial gain, whether in pharmaceuticals, retail, finance, or any other sector, can take inspiration from how this technology accelerates discovery and sharpens strategic focus. The same capabilities that help researchers avoid dead ends could help businesses streamline decision-making, predict market trends, and personalise offerings with remarkable precision.

However, as with any transformative technology, there is a need for cautious optimism. Ethical considerations, potential job displacement, and the risks of over-reliance on AI should not be overlooked. The key will be fostering a collaborative relationship between human expertise and machine intelligence, much like the partnership between Imperial College London’s researchers and Google’s AI tool.

Looking ahead, the real triumph will come from how effectively industries and institutions integrate AI into their workflows, not as a replacement for human creativity but as a co-pilot that enhances our ability to solve problems. For both science and business, this breakthrough could represent not just a faster path to solutions, but an entirely new way of thinking about what’s possible when human ingenuity meets machine precision.

Microsoft has unveiled Majorana 1, the world’s first quantum chip powered by a ‘Topological Core architecture’, which it claims could enable quantum computers to solve complex, industrial-scale problems within years rather than decades.

The Issue

The Majorana 1 chip could signify a pivotal shift in quantum computing development. Unlike conventional processors, which rely on classical bits (the familiar ones and zeroes of modern computing), quantum computers use qubits, i.e. quantum bits that can represent both states simultaneously. While this promises an exponential increase in processing power, qubits are notoriously difficult to stabilise and control due to environmental interference.

Microsoft’s Revolutionary Approach to Quantum Architecture

In the case of Microsoft’s Majorana 1, instead of relying on traditional qubit designs, the company has taken a more ambitious route by developing a new material called a topoconductor. This breakthrough enables the manipulation of elusive Majorana particles, which were once purely theoretical and only recently demonstrated in laboratory conditions.

The creation of this topological state of matter, a new form distinct from solids, liquids, or gases, has therefore allowed Microsoft to produce topological qubits. The advantage is that these are expected to be more stable, less prone to error, and capable of being controlled digitally rather than through complex analogue mechanisms.

Years Rather Than Decades

Chetan Nayak, a technical fellow at Microsoft, has explained the significance of the innovative technology used in the new chip, saying: “Many people have said that useful quantum computers are decades away. I think that this brings us into years rather than decades.” This optimism appears to be built on the company’s ability to scale its technology, aiming for an unprecedented one million qubits on a single chip.

Industrial-Scale Problems Within Reach

The potential impact of this innovation could be transformative across industries. Quantum computers have the capacity to simulate molecular interactions, design new materials, and solve optimisation problems that would take today’s most powerful supercomputers millions of years to process. Microsoft believes these capabilities could unlock advancements in:

– Pharmaceuticals. Accelerating drug discovery by simulating molecular structures with unprecedented precision.

– Energy storage. Designing better, more efficient batteries for electric vehicles and renewable energy.

– Environmental solutions. Developing catalysts to break down microplastics or reduce carbon emissions.

– Advanced manufacturing. Creating self-healing materials for infrastructure, reducing maintenance costs and enhancing safety.

A New Front in the Quantum Computing Race

Microsoft’s announcement about its new chip will, no doubt, have sent ripples across the already competitive quantum technology landscape. Rivals such as Google and IBM have made significant strides with quantum processors using alternative qubit designs. Google’s “Sycamore” processor, for example, made headlines in 2019 for achieving quantum supremacy by solving a problem in 200 seconds that would take classical computers 10,000 years. However, Microsoft’s strategy, though slower in producing short-term results, may prove more scalable in the long run.

While Microsoft’s prototype currently houses eight topological qubits, far fewer than the hundreds achieved by competitors, the company’s promise of a clear path to a million qubits sets it apart. However, experts believe that if Microsoft’s technology can indeed scale as planned, it could actually leapfrog its rivals in the race to build commercially viable quantum machines.

Business and Industry

For businesses and industries poised to embrace quantum computing, this development could radically shift the landscape. For example, being able to solve industrial-scale problems within years rather than decades could lead to:

– Faster innovation cycles. Products designed and tested virtually with quantum precision could dramatically reduce time-to-market.

– Cost reductions. More efficient materials and manufacturing processes could slash production costs.

– Sustainability breakthroughs. Quantum modelling could enable the development of eco-friendly materials and more efficient energy solutions.

Accessing Quantum Capabilities Through The Cloud

Microsoft’s integration of the Majorana 1 chip into its Azure Quantum platform means that businesses will eventually be able to harness these capabilities through cloud services, thereby democratising access to quantum power without the need for prohibitively expensive infrastructure.

A High-Risk, High-Reward Strategy

Microsoft’s focus on topological qubits appears to have been quite a high-risk strategy, given the scientific and engineering challenges involved. For example, until recently, Majorana particles had never been observed in nature and had to be coaxed into existence through precise manipulation of materials at the atomic level.

However, as Krysta Svore (another Microsoft technical fellow) pointed out, the architecture’s simplicity could allow for rapid scalability. Svore said: “It’s complex in that we had to show a new state of matter to get there, but after that, it’s fairly simple. You have a much simpler architecture that promises a faster path to scale.”

The Next Steps for Quantum Computing

Microsoft’s inclusion in the US Defence Advanced Research Projects Agency’s (DARPA) Underexplored Systems for Utility-Scale Quantum Computing (US2QC) programme signals the strategic importance of this technology. If successful, the company could deliver the world’s first utility-scale, fault-tolerant quantum computer, a machine whose computational value exceeds its operational costs.

For now, though, the road ahead remains fraught with technical challenges. Scaling from eight qubits to a million will require solving issues of coherence, error correction, and manufacturing precision on an unprecedented scale.

That said, if Microsoft’s bet pays off, the promise of solving industrial-scale problems within a matter of years could mark the beginning of a new technological era, one where quantum computing transforms everything from materials science to global sustainability efforts.

What Does This Mean For Your Business?

Microsoft’s unveiling of the Majorana 1 chip represents a potential shift in the trajectory of quantum computing itself. The company’s bold move to pursue topological qubits through the manipulation of Majorana particles looks like being both an audacious scientific gamble and a forward-thinking strategy aimed at overcoming some of the most persistent obstacles in the field.

While rivals like Google and IBM have made headlines with short-term achievements using more traditional qubit designs, Microsoft’s approach seeks to tackle the longer-term challenge of scalability and stability. By leveraging a fundamentally different quantum architecture, the company may ultimately sidestep the fragility that plagues conventional quantum systems. If successful, this could place Microsoft at the forefront of a technological race that has, until now, seemed more theoretical than practical.

It should be noted that, although the signs are good, caution is needed because technical hurdles like maintaining coherence and error correction are not trivial and could be pretty challenging for Microsoft. That said, Microsoft’s confidence, underpinned by integration with its Azure Quantum platform, suggests a readiness to bring quantum capabilities to businesses and researchers sooner than previously imagined.

The implications for industry and society at large could be transformative. From revolutionising drug discovery to enabling breakthroughs in clean energy and sustainable manufacturing, the possibilities of scalable quantum computing extend far beyond academic curiosity. The prospect of solving industrial-scale problems in years rather than decades could accelerate innovation cycles, reduce costs, and unlock sustainable solutions previously out of reach.

Microsoft has reported uncovering a cyberattack campaign by Storm-2372, a group linked to Russian interests, using a technique called device code phishing to bypass multi-factor authentication (MFA) and steal access tokens.

Active since August 2024, the group targets governments, NGOs, and industries including defence, telecoms, energy, and healthcare across Europe, North America, Africa, and the Middle East. In device code phishing, attackers trick users into entering a legitimate authentication code, sent via fake meeting invites on platforms like Microsoft Teams and WhatsApp, on a genuine sign-in page. This hands over valid tokens, granting unauthorised access.

Recent activity shows a shift towards using Microsoft Authentication Broker’s client ID to gain persistent access by registering rogue devices inside compromised networks. Microsoft warns these attacks are especially effective because they mimic legitimate login workflows.

To defend against device code phishing, businesses should block unnecessary device code flows, strengthen Conditional Access policies, educate users about phishing risks, and use phishing-resistant MFA methods such as FIDO tokens.

Scientists from a Swiss university have discovered a way to break down Plexiglass into its original building blocks using violet light and a common solvent, thereby making recycling plastics far more efficient and potentially helping to tackle global plastic waste.

Cracking the Plastic Code: The Science Behind the Discovery

The process, developed by lead researcher Dr Hyun Suk Wang at ETH Zurich, works by exposing Plexiglass, a type of polymethacrylate, to violet light while it is submerged in dichlorobenzene solvent. The scientists discovered that this exposure releases chlorine radicals from the solvent, which then break apart the strong carbon-carbon bonds in the plastic. The result is the recovery of methyl methacrylate (MMA), the original monomer building blocks from which Plexiglass is made.

This recovered monomer can be purified and repolymerised without losing any material quality, unlike traditional recycling methods that involve shredding, cleaning, and remelting. Those older methods degrade the properties of plastic with each cycle, whereas this new chemical process allows the material to be fully restored to its original state.

The Scale of the Plastic Waste Problem

The scale of plastic pollution globally remains a significant challenge. For example, over 400 million metric tonnes of plastic waste are produced worldwide each year and yet, only around 9 per cent of this waste is successfully recycled. Also, rather than being recycled, half ends up in landfills, while another 19 per cent is incinerated. One particularly damaging effect of our plastic use is that around 11 million metric tonnes of plastic enter the ocean annually, harming ecosystems and marine life.

Plexiglass Particularly Problematic

Polymethacrylates like Plexiglass are particularly problematic due to their durability and widespread use in industries ranging from construction to electronics. This resilience, while useful in manufacturing, makes them resistant to breaking down in traditional recycling systems.

Closing the Loop

Lead researcher Dr Hyun Suk Wang and his team have said they believe their light-based method could transform how Plexiglass and similar plastics are recycled. Dr Wang says: “By recovering monomers in near-pristine condition, we can effectively close the loop on Plexiglass production.”

The Implications

If adopted at scale, the implications of this breakthrough could include:

– Reduced use of fossil fuels. Since virgin plastic production depends on fossil resources, recycling monomers could significantly cut demand for petrochemical feedstocks.

– Lower energy consumption. The process requires less energy than current methods, which often involve high temperatures and extensive mechanical processing.

– Industrial adaptability. Preliminary tests suggest that the process can be applied on a larger scale with precision and control, making it a candidate for industrial recycling operations.

Is It Scalable?

It should be noted, however, that for this discovery to be commercially viable, several key challenges need to be addressed, which include:

– Being able to generate violet light at scale. The process depends on specific wavelengths of light, meaning industrial-level violet light sources would be necessary.

– Handling dichlorobenzene safely. The solvent used in the process is hazardous and would require strict handling protocols to ensure worker and environmental safety.

– Economic feasibility. Any new technology must be cost-competitive with the low expense of producing virgin plastics from petrochemicals.

Despite these hurdles, the researchers remain optimistic. As co-author Professor Athina Anastasaki points out, “What makes this process so promising is its ability to work on a wide range of polymethacrylates, regardless of how they were originally manufactured.”

What Next?

The research team is now working on refining the technique to handle mixed plastic waste streams, a major obstacle in current recycling systems. They are also exploring alternative, less toxic solvents to improve the process’s environmental impact.

At the same time, discussions are taking place with industrial partners to assess how this technology might be integrated into existing recycling facilities.

What Does This Mean For Your Organisation?

This breakthrough in recycling Plexiglass using violet light and a common solvent could mark a promising step forward in addressing the global plastic waste crisis. The discovery by Dr Hyun Suk Wang and his team at ETH Zurich presents a genuinely innovative approach – one that allows plastics to be broken down into their original building blocks without degrading their quality. By recovering monomers in a near-pristine state, this method could redefine what it means to “recycle” plastics, moving beyond the traditional processes that weaken materials with each cycle.

The potential environmental benefits are clear. If this technology can be successfully scaled, it could significantly reduce the dependence on fossil fuels required for producing virgin plastics, cutting both carbon emissions and petrochemical consumption. Furthermore, the process’s lower energy demands compared to conventional recycling could provide a more sustainable and economically viable solution, particularly for industries with high energy consumption rates.

For businesses, especially those in manufacturing, construction, and consumer goods, this development could offer both economic and strategic advantages. Companies that rely heavily on plastics might see reduced costs in sourcing high-quality recycled materials, avoiding the need to purchase more expensive virgin plastics. Also, integrating this technology into supply chains could help businesses meet increasingly stringent sustainability targets and regulatory demands around recycling and carbon emissions.

Beyond compliance, there is also the potential for businesses to strengthen their brand reputation by aligning with environmentally responsible practices. Early adopters of such groundbreaking recycling methods could position themselves as leaders in sustainability, attracting eco-conscious consumers and investors alike. However, industries will need to assess the commercial feasibility carefully, considering factors such as the cost of installing violet light technology and handling hazardous solvents like dichlorobenzene.

That said, significant obstacles remain. The need for scalable violet light sources and safe handling of potentially hazardous solvents are non-trivial challenges that could slow widespread adoption. Also, the economic viability of this method will need to be thoroughly tested against the low costs associated with producing virgin plastics, a factor that has historically undermined efforts to expand plastic recycling.

The optimism shown by researchers like Professor Athina Anastasaki highlights the broader potential of this technology. If successful refinements are made, particularly in handling mixed plastic waste streams and identifying safer solvents, the process could become adaptable enough for industrial-scale use.

While this innovation is not without its hurdles, this research looks as though it could open an exciting new chapter in the fight against plastic pollution. If industry stakeholders, policymakers, and scientists can work together to overcome the technical and economic barriers, this light-driven recycling method could play a pivotal role in creating a truly circular economy for plastics.

This video explains how to use data that’s publicly available to enable your AI of choice to suggest those topics for social media that your audience is most likely to engage with.

[Note – To Watch This Video without glitches/interruptions, It may be best to download it first]

You don’t need to stop your ChatGPT conversations just because you switch apps or lock your phone – there’s a handy feature to keep the chat flowing in the background. Here’s how it works:

How to:

– Open the ChatGPT app and tap the two-line menu button in the top-left corner.

– Press your account name at the bottom of the menu to access settings.

– Select ‘Voice’ and enable ‘Background Conversations’.

– Now, you can talk to ChatGPT while using other apps or even when your screen is locked.

– To disable this feature, simply turn off ‘Background Conversations’ from the same menu.

Bonus:

– On the same screen, you can customise the assistant’s voice. Tap ‘Voice’, scroll through options, and pick your favourite (e.g. Vale and Arbor are British ones).

– Hit Done to save your preferences.

This is perfect for multitasking, like asking questions while browsing or replying to messages on WhatsApp without pausing your conversation.