Plastics, oil, petrol– the modern world is entirely dependent on hydrocarbons. The good sources are slowly running low and supply is increasingly complicated by geopolitical factors we really don’t want to get into, but hey! It’s just hydrogen and carbon, right like it says in the name. How hard could it be to roll your own at home. Well, if you’ve got a lab like [Marb]’s Lab on YouTube, it might just be doable, as he demonstrates in his latest video.

The Fischer-Tropsch reaction was discovered back in 1925 in Germany by a couple of gents named Fischer and Tropsch. In the unpleasantness that followed later, Germany made good use of their process on an industrial scale, since they had ample coal and no oil on hand. Coal-rich South Africa has also made us of it, particularly during the Apartheid-era trade restrictions. Every so often the idea of industrializing the process comes up in the USA, but there’s still enough oil there it doesn’t make sense economically.

Those nations all have something in common: they’re all coal-rich countries, and that makes sense because coal is easily converted to carbon monoxide and hydrogen– a combo known as syngas– and it just so happens that those are the feedstock for this reaction. The actual chemistry going on inside is quite complex, but conceptually it is pretty simple: hydrogen and carbon monoxide mix over a hot metal catalyst, and combine to form various hydrocarbons.

In [Marb]’s glassware-based demonstration, the catalyst is Cobalt (III) Oxide on silica gel– a lovely, cancer-causing substance that must be prepared for each use, as it lasts but 24 hours before further oxidization ruins it. That’s in spite of purging the system with argon– a necessary step if one does not wish to explode. The yield isn’t amazing, and [Marb] isn’t sure exactly what mix of hydrocarbons he has created– although they smell like gasoline and burn like the dickens, so mission accomplished.

This might seem like the furthest thing from green, but if you use solar power to run the process and something like woodgas– which is syngas by any other name– as a feed-stock, then you’ve got a carbon neutral energy storage medium.

Thanks to [Markus Bindhammer] for the tip!

from the hallucinating-your-arrest dept

This article is republished from The Conversation under a Creative Commons license. Read the original article.

In Baltimore County, Maryland on Oct. 20, 2025, a 17-year-old student named Taki Allen was sitting outside his high school after football practice when an artificial intelligence-enhanced surveillance camera falsely identified the Doritos bag in his pocket as a gun. Within moments police cars arrived, officers drew their weapons and Allen was forced to his knees and handcuffed while they searched him. All they found was a crumpled bag of chips. The AI’s misidentification and the human decisions that followed turned a normal evening into a traumatic confrontation.

On Dec. 24, 2025, Angela Lipps, a Tennessee grandmother, was released after spending five months in jail because facial recognition software had incorrectly connected her to fraud crimes in North Dakota, a state she had never visited. Police had arrested her at gunpoint while she was babysitting her four grandchildren.

These are unfortunate examples of how AI can lead to mistreatment of people because of technical flaws as well as misplaced human faith in the technology’s supposed objectivity. These cases involve different tools, but the underlying issue is the same. AI systems produce probabilities, and people treat them as certainties.

We are researchers who study the intersection of technology, law and public administration. In researching how police departments use AI and how digital technologies operate in a democratic society, we have seen how quickly the shift from probabilistic prediction to operational certainty happens in practice.

AI policing tools are used in dozens of U.S. cities, although no public registry tracks the full footprint. The tools ingest historical crime data and score neighborhoods on predicted risk so officers can be routed toward the resulting hot spots. The mechanism is straightforward, but its consequence is not. Once a system signals a possible threat, the question is no longer how certain the prediction is but what to do about it. A statistical output turns into a deployment decision, and the uncertainty that produced it gets lost on the way.

A matter of probabilities

When generative AI models such as ChatGPT or Claude respond to human requests, they are not searching a database and pulling out facts. They are predicting the most likely answer based on patterns in data they have been trained on. When asked, “Who invented the light bulb?” the models do not go to a source or fact-check a finding. They generate a statistically probable answer which is “Thomas Edison.” The reply might be right, but it might not capture the full story – such as Joseph Swan’s parallel invention at the same time as Edison’s. The danger arises when people believe that the model is retrieving truth rather than generating likelihoods.

This distinction matters. The most probable response is not the same as a factually verified answer, complete with context.

This reality can be highly problematic for policing and law. For example, when law enforcement agencies use AI systems trained on geographical data to estimate where criminal activity is likely to occur, the algorithms analyze historical crime data and geographic patterns. These systems generate statistical risk scores or heat maps for locations based on prior incidents. But such predictions may have little bearing on who was involved in a new crime in the area, even if an algorithm generates information that sounds authoritative.

Some researchers have argued that predictive policing systems do not increase the likelihood that racial minorities will be arrested more often relative to traditional policing practices. The broader concern, however, is not limited to measurable disparities in arrest outcomes alone. It is about how probabilistic predictions can become standardized operational decisions absent further verification.

Artificial intelligence researchers caution against using these models in isolation for crime and legal proceedings or decision-making. Research at the University of Virginia’s Digital Technology for Democracy Lab with police chiefs shows that some law enforcement groups follow strict policies that dictate when technology is used in tandem with, or in place of, human discretion, while others have no such policy.

What most users do not realize is that AI systems rarely produce binary answers: yes or no, a positive identification or a negative one. They generate probabilities. Some systems assign scores that assess the system’s confidence in a prediction. In those cases, engineers set a confidence threshold, a level of certainty that determines when the system should trigger an alert about a possible threat. You can think of this threshold as settings on a control knob. A 95% confidence level, for example, indicates that the model considers its interpretation to be highly likely.

A low threshold catches more potential threats but increases false alarms. A high threshold reduces mistakes but risks missing real dangers. Either way, these algorithmic thresholds are often invisible to the public and are set quietly by vendors or agencies, even though they shape when police action begins.

Where to draw the line

In medicine, these kinds of trade-offs are explicit. Diagnostic tools are calibrated on the relative harm of different errors. In infectious disease settings, for instance, systems that detect infections are often designed to accept more false positives to avoid missing contagious individuals. Then medical professionals look into the human cases. And the algorithm-based decisions are subject to professional standards, ethics reviews and regulatory oversight.

In policing, an AI system must balance false positives, where the system flags a threat that does not exist, and false negatives, where it fails to detect a real danger. The trade-off carries significant consequences. A lower threshold may generate more alerts and allow officers to intervene earlier, but it also increases the risk of mistaken identifications, which happened to Angela Lipps, or escalated encounters like the one Taki Allen experienced. A higher threshold may reduce wrongful interventions but could allow legitimate threats to go undetected.

Some law enforcement agencies argue that acting on imperfect signals is preferable to missing serious risks. But lowering the bar for algorithmic alerts based on probabilistic estimates effectively expands the number of people subjected to police attention. It is important to realize that these thresholds are not neutral features of the technology; they are choices embedded by the creators in the model’s code. Decisions about where to draw the line determine when an algorithmic suspicion becomes a real-world police action, even though the public rarely sees or debates how those thresholds are set.

Limits of optimization

Developers often use several methods to determine where to set a confidence threshold. Techniques such as “receiver operating characteristic curve analysis” examine how changing the threshold for an alert alters the balance between correctly identifying real events and mistakenly flagging harmless ones. Precision–recall analysis examines a similar trade-off, asking how accurate the system’s alerts are relative to the number of incidents it successfully detects.

These approaches could help calibrate systems more responsibly by testing how often an algorithm wrongly flags people or locations. Fine-tuning can improve system performance. But the techniques cannot resolve the underlying question of how much algorithmic uncertainty society is willing to tolerate.

In law, legal standards of proof determine how convincing evidence must be before a judge or jury can rule in favor of a plaintiff or defendant. Courts use formal standards of proof depending on the stakes, such as probable cause, preponderance of the evidence and beyond a reasonable doubt. These standards reflect a societal judgment about how much uncertainty is acceptable before exercising legal authority. A court does not accept a guess or a prediction; it follows a process to weigh evidence. Unlike humans, an AI model does not usually say, “I’m not sure.” A model typically has confidence in its reply, even when the answer is incorrect.

Stakes are rising as AI enters the courtroom, law enforcement, the classroom, the doctor’s office and the public sector. It is important for people to understand that AI does not know things the way many assume it does. It does not distinguish between “maybe” and “definitely.” That is up to us. We believe that technologists should design systems that admit uncertainty and need to educate users about how to interpret AI outputs responsibly.

Maria Lungu is a Postdoctoral Researcher of Law and Public Administration at University of Virginia and Steven L. Johnson, is Associate Professor of Commerce at University of Virginia

Filed Under: ai, arrests, policing, wrongful arrest

Sidephone has built an Android phone around a swappable USB keypad, entering a niche between the distraction-heavy smartphone and the stripped-back dumbphone by offering physical input options without sacrificing app compatibility.

We saw the device at Beyond Expo 2026, where its keypad system came in two layouts, a QWERTY configuration for text-heavy users and a numpad layout for those who prefer a more traditional input style, with a growing selection of cases and colour options.

That restraint extends to the camera, which prioritises simplicity over the high-resolution computational photography that dominates modern flagships, producing shots designed to be printed and kept rather than shared online.

Running Android without Google or Google Play services, the device supports thousands of existing apps that function independently of Google’s infrastructure, covering essentials like calls, SMS, camera, files, and a gallery app out of the box.

Advertisement

Advertisement

Privacy-focused app ecosystem

Beyond the defaults, Sidephone ships with several optional app bundles focused on privacy and open-source software, including the full Proton suite covering mail, VPN, calendar, drive, and password management, as well as Signal, WhatsApp, and a browser choice between Vivaldi and Firefox.

The Essentials bundle adds Maps powered by HERE, LocalSend for cross-platform file sharing, a local music player via Auxio, and a voice recorder, while the Misc bundle extends further with AntennaPod for podcasts, BreezyWeather, and Obtanium for installing apps directly from developer release pages.

Support for third-party messaging apps, including Telegram, Threema, and WeChat, is also listed as a work in progress, alongside compatibility with music streaming services such as Apple Music, Spotify, and Tidal.

For anyone considering making the switch, more information and ordering are available on the Sidephone website, with the broader community active on Reddit at r/Sidephone.

Advertisement

• The same public ledger that enables transparency in crypto often acts as a double-edged sword for some of its whales, who are identified and targeted by hackers, con artists, and other criminal elements
• Bloomberg is reporting a 75% increase in recorded physical attacks (also known as crypto wrench attacks) against cryptocurrency holders year-on-year in 2025
• Whales, crypto-related firms, and exchanges have responded by upping the ante on security protocols, increasing bodyguards, and even employing pre-emptive measures

Cryptocurrency executives and whales alike are increasingly being targeted by a mix of criminal elements worldwide, even as security continues to be beefed up to protect the not-so-anonymous owners of cryptocurrency.

The transparency introduced to the crypto world is putting some coin-collectors at risk of physical harm, and even kidnapping.

Kyndryl said the ‘politicisation’ of the deal has ‘overshadowed’ potential benefits it could have provided to Dutch citizens.

The government of the Netherlands has blocked Kyndryl’s proposed acquisition of Dutch cloud provider Solvinity, citing a risk to public interest.

Solvinity stores data used by the country’s citizen identification tool DigiD. The company’s tools are also used by public institutions including the tax office and universities. The acquisition would have reportedly set Kyndryl back €100m.

“The [Dutch] Investment Screening Bureau (ISB) advised me to proceed with a complete prohibition of this acquisition. I have made this advice my own and have adopted it,” said Willemijn Aerdts, the country’s minister for the digital economy and sovereignty, in a translated letter to parliament.

“The Netherlands attaches great value to the presence of foreign, including explicitly American, technology companies and their contribution to the Dutch economy and digital infrastructure,” the minister added, though she did not explain why the acquisition would risk public interest.

This is the first time the ISB has blocked a US acquisition since it was set up in 2020.

Several members of the Dutch parliament had openly criticised Solvinity’s acquisition by a US IT company over concerns that the US government could access private European citizen data.

“We are extremely disappointed by the Netherlands’ government decision to prohibit Kyndryl’s acquisition of Solvinity,” said Kyndryl in a statement.

“Since announcing the proposed transaction, Kyndryl has consistently engaged in good faith with relevant stakeholders across the Netherlands’ government.

“Despite this engagement and our long history of managing mission-critical operations in the Netherlands, the politicisation of this process has overshadowed the clear and important benefits this transaction would have brought to Solvinity’s customers and Dutch citizens”.

Kyndryl announced the proposed acquisition last November, stating that the deal would enable it to offer customers expanded services in “modernising, innovating and securing sensitive and complex workloads”.

Meanwhile, last September, the Dutch government began seizing Chinese-owned chipmaker Nexperia, citing a “threat” to Europe’s semiconductor capabilities.

The back-and-forth led to China temporarily halting Nexperia chip exports in early October. However, a Dutch court later upheld the seizure decision, as well as a decision to suspend the company’s Chinese CEO and hand control off to EU-based directors.

The dispute, triggered by the Dutch government, led to parent company Wingtech suing its subsidiary Nexperia earlier this month, arguing that the government’s actions resulted in billions in losses. The company is demanding €1bn in damages.

Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.

Ivan Miranda has turned a long-running experiment with rolling balls into a clock that actually keeps pace with passing seconds. The latest version of his marble timepiece refreshes its display fast enough to show hours, minutes, and seconds without any visible lag between changes. Each digit takes shape inside a compact 3-by-5 grid. Fifteen marbles settle into the exact spots needed to outline a number. White marbles stand where a digit needs its bright segments, while black marbles fill the remaining positions to create clear contrast. The result looks like a physical version of familiar numeric shapes, built from actual objects rather than light or ink.

Hundreds of marbles, a stunning collection, live at the bottom of the machine, waiting to be sorted, and to be honest, they are getting a good exercise there. So, an elevator moves rows of marbles up in batches at a time, and infrared sensors examine each marble’s colour as it rises. Colors that do not match are softly shoved to one side by little actuators, and the entire cycle is repeated. They merely keep going until they have the perfect marble mix, at which point the mix required for the next exhibit is tossed down the tight tracks. Gravity takes over, transporting the picked marbles to their proper spots in the grids.

For years, the machine simply chugged away, updating once per minute. It was just the right pace to allow the marbles to organize themselves, travel to their destination, and settle in. However, adding only a few seconds to the mix nearly threw everything off course. Marbles were jamming up against the walls or each other, black and white marbles would take slightly different routes, causing them to be out of sync with one another, the plastic parts would flex and bend under the loads, and the drive system was nearing failure due to the large spikes in power demand.

Miranda needed to think of a new way to do things, so he developed a series of modifications that all worked together. He expanded the main housing so that marbles could move around without becoming too crowded. Then he built a safety net, which is a type of buffer zone that collects all marbles for a digit in one location before the last move. Release is now accomplished using gates carved from robust aluminum and equipped with a thin metal plate that serves as both a hinge and a spring. The flywheel, which weighs just shy of two kilograms, simply spins silently away on the drive shaft, keeping the motion flowing smoothly even as the mechanism begins to pull strongly. They took sure to include a great smooth-riding joint in the axle so that power is properly distributed even though the layout results in an unusual angle.

All of these modifications resulted in the 15 marbles for each digit dropping into their final positions in around 150 milliseconds. Which is just fast enough for the display to change once per second while still producing clean, crisp numerals. The time is handled by an Arduino controller, which is connected to a constant clock source and is in charge of communicating with the sensors and actuators. The motors and gears handle all of the heavy lifting, with the flywheel smoothing out any peaks and troughs that would otherwise cause the entire system to stall.

Recycling is still the slowest part, as the marbles tumble back down to the collection area after each cycle and must be hoisted again for the next cycle. The existing elevator just cannot match the one-second pace for lengthy run times, therefore the equipment continues to hesitate or slow down during long tests. Even with such limitation, the core display is able to process updates at a rate comparable to real-life seconds.

The entire design is a bit of a hybrid, as he used 3D printed tracks and frames where they made sense, and only a few proper machined metal parts where they were truly needed. Miranda has also been documenting every step of the route on camera, highlighting the worst of the clogs that slowed progress and the best of the solutions that restarted the flow. What began as a ridiculous idea about moving actual pixels has evolved into a workable proof of concept for how gravity, sensors, and properly formed pieces can accomplish tasks such as spitting out correct seconds without the traditional bits, hands, or glowing bits.
[Source]

The Global Tech Ecosystem Index for 2026 tracks start-up ecosystems in 325 cities across 77 countries through three categories.

European cities account for 10 of the top 20 densest tech ecosystems in the world, according to a new report by innovation tracking platform Dealroom.

The ‘Global Tech Ecosystem Index’ for 2026 tracks start-up ecosystems in 325 cities across 77 countries through three categories: scale, per capita performance – or density – and growth, aiming to provide a “multidimensional view of the global innovation landscape”.

In density ratings, 45 European cities feature in the global top 100, ahead of North America’s 40. The rankings measure innovation output per capita, including start-up activity, enterprise value creation, ‘unicorns’ and university affiliations.

Cambridge, London, Stockholm, Ghent, Lausanne, Oxford, Tallinn, Copenhagen, Munich and Amsterdam are among the 20 global density leaders. The UK’s Cambridge is third in this metric, behind only the Bay Area and Boston, both in the US.

US ecosystems account for nine entries overall for density leaders.

Europe is one of six global macro regions monitored by the index – alongside North America; Latin America; Asia-Pacific; Middle East and North Africa; and Sub-Saharan Africa – for economic ties, investment flows and tech ecosystem integration.

The index suggests this approach reflects how innovation ecosystems operate in practice, connected by trade, talent and capital rather than by strict geography alone.

In the index’s scale rankings, tracking the “world’s largest and most successful tech ecosystems”, London (fourth), Paris (eighth) and Stockholm (19th) feature in the top 20, with 11 places occupied by North American cities.

Istanbul, Zagreb and Kyiv are European entries in the top 20 growth locations globally.

“As Europe struggles with entrenched low growth, political leaders see tech and innovation as a fundamental building block of economic competitiveness,” said Yoram Wijngaard, Dealroom’s founder and CEO.

“Innovation is seen as a route not only to securing growth but also national resilience and strategic autonomy. What stands out is not just the strength of leading hubs like London and Paris, but the rise of high-performing smaller ecosystems often built around leading research and academic institutions.”

In the AI sector, European ecosystems account for 10 of the index’s global top 20 by density, with Cambridge ranking second for the category. Tel Aviv, Israel and nine US cities complete the category.

In the index’s defence sector top 20 by density, European cities account for six entries, with Munich ranking second for the category.

Dealroom said: “The rankings reflect Europe’s distinctive innovation model – globally competitive ecosystems built around research institutions, technical talent and specialised industries rather than scale alone. Across the continent, smaller cities are increasingly producing outsized impact in areas such as AI, biotech, climate-tech and advanced manufacturing.”

According to Dealroom, its database draws from four sources: aggregated public information such as news, filings, registries, job boards; community-submitted data from start-ups, investors and accelerators validated by Dealroom; direct government API connections; and third-party data partnerships.

Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.

Looking for a portable Bluetooth speaker that will not disappoint in terms of performance? The Ultimate Ears WONDERBOOM 4, priced at $51.44 (was $100), is an ideal fit. It fits neatly into daypacks or backpacks without adding weight, yet produces music that penetrates through even small spaces and outdoor gatherings with startling power.

This speaker is 4 inches tall and weighs 15 ounces, making it easy to pick up and move with one hand, whether you’re setting it up or simply repositioning it. On top, there is an elastic loop that can be clipped into your backpack or bike handle without bulking up. The device is coated in a fabric finish and has some excellent rubberized portions at the ends to keep it gripping even when your hands are damp and soggy. Plus, it’s made to withstand some punishment: the IP67 rating means you can put it in a muddy puddle and still know it’ll work, and the fact that it floats means that if you drop it in a pool or lake, it’s less of a nightmare and more of a ‘oh well, I’ll just fish it out’.

Sale

Ultimate Ears WONDERBOOM 4 Portable Waterproof Bluetooth Speaker with Balanced 360-Degree Sound and Bass…

• Balanced 360-Degree Sound: The WONDERBOOM 4 portable waterproof Bluetooth speaker features 360-degree sound in a petite package; boosted sound for up…
• Outdoor Boost and Podcast Mode: Engineered for full stereo sound; tap the Outdoor Boost button for outdoor environments; use the new Podcast Mode to…
• 14 Hours of Boom: With a rechargeable battery that lasts throughout the day, this waterproof portable speaker booms on with up to 14 hours of…

The sound is, frankly, pretty impressive all things considered. See, the configuration of the two active drivers and some passive radiators is set up such that it’s nice and open, which is wonderful news for anyone who wants to use it from a table or on the ground, and it works nicely no matter how you hold it, so you don’t have to worry about setting it perfectly. The bass, while not ridiculously deep, has a great weight to it, and the vocals are audible whether you’re listening to a podcast or just chilling out to some music.

There are also some extremely cool features to look out for. For example, tapping the button at the bottom activates the Outdoor Boost feature, which helps cut through background noise and get the sound through loud and clear. You can even pair a second WONDERBOOM 4 to get proper stereo separation, which is a game changer for larger rooms or sharing with a friend because it provides a considerably wider soundscape without requiring any additional cords or downloading anything. It also has multipoint Bluetooth connections, which allow you to couple with two devices at the same time and effortlessly switch between them.

The controls are simple to use, with volume buttons right up front for quick adjustments on the go. Everything else, including power, pairing, and playback controls, is conveniently located up top. The good news is that you don’t have to be a tech whiz to get started; simply plug it in, flip the switch, and you’re ready to go. You’ll have up to 14 hours of gameplay, which should provide for a road trip, a few shorter sessions, or a full day of work, plus some evening listening before you need to recharge. When you’re done, simply grab a standard USB-C cable to get back up and running.