Around one in three Cypriots say they would use the digital euro in their daily lives, despite limited awareness of the new form of money, according to the first islandwide survey published by the Central Bank of Cyprus.

With the first issuance currently expected in 2029, the findings suggest that public education will be crucial, particularly among people who rely more heavily on cash or have less experience with digital tools.

Follow THE FUTURE on LinkedIn, Facebook, Instagram, X and Telegram

Awareness Remains Low

Some 61% of respondents say they have no knowledge of the digital euro, while just 1% consider themselves fully informed.

Awareness is higher among people under 65, those with tertiary education and employed respondents. Among those who have heard of the digital euro, awareness is also more common among men, higher-income and more highly educated people, as well as urban residents.

Social media is the leading source of information, cited by 49% of respondents, followed by television at 30%.

35% Would Use The Digital Euro

Despite the knowledge gap, 35% say they are willing to use the digital euro in their daily lives. This is particularly true among people under 45, employed respondents and those with higher education and incomes.

Among potential users, 41% would use it for purchases in physical shops, 40% for online shopping and 33% for person-to-person payments.

By comparison, 28% say they are somewhat or very unlikely to use the digital euro.

Privacy And Cash Are Main Concerns

The biggest concerns are the possibility of transactions being tracked and fears that cash could eventually be abolished, cited by 53% of respondents.

Another 38% are concerned about security, while 25% worry about managing their spending. Some 30% have significant concerns about the ease of using the digital euro.

For businesses, 9% say their willingness to accept digital euro payments would depend on factors such as cost, ease of implementation and demand, while 27% say they would not accept such payments.

OpenAI has introduced new safeguards for developing and testing advanced AI models, including stronger monitoring, tighter network isolation and additional security checks during post-training.

The company said the measures are intended to keep pace with growing risks as AI systems become more capable. OpenAI also said the changes were influenced by the cybersecurity capabilities expected from its upcoming Astra model, as well as the broader acceleration of AI development.

Follow THE FUTURE on LinkedIn, Facebook, Instagram, X and Telegram

Changes Follow Hugging Face Incident

The new measures come weeks after OpenAI disclosed a security incident involving Hugging Face, where models escaped their training environment after an internet-connected tool on OpenAI’s network was compromised.

OpenAI said the new policies are not a direct response to the incident, but acknowledged that it had paused reinforcement learning for two weeks afterward. Less risky training has since resumed, while its largest planned frontier reinforcement learning run remains on hold pending further testing and safety evaluations.

Stricter Monitoring For Advanced Models

OpenAI said its security requirements will become stricter as models become more capable, with the most advanced systems receiving the highest level of scrutiny.

A new monitoring system will track tool actions, available reasoning traces and activity logs to detect potentially unauthorized behavior. The company aims to generate alerts within 30 minutes of suspicious activity.

The monitoring is expected to require computing resources equivalent to around 20% of the process being monitored.

Stronger Network Isolation

OpenAI is also introducing tighter network controls designed to prevent a single compromised system or service from gaining access to the internet or other internal networks.

The company has not yet provided detailed technical information about the new safeguards. It also said a full postmortem of the Hugging Face incident is still pending.

Amazon plans to expand its Prime Air drone delivery service to nearly 500 U.S. cities and towns by the end of 2026, a sixfold increase from its current footprint. The company says it has already delivered hundreds of thousands of packages by drone this year, with thousands of deliveries made daily.

Prime Air Expands After Years Of Delays

Amazon first unveiled its drone delivery vision in 2013, promising deliveries within 30 minutes. Since then, the program has faced regulatory delays, technical problems and opposition from some communities.

Follow THE FUTURE on LinkedIn, Facebook, Instagram, X and Telegram

A major step forward came in 2024, when Amazon received regulatory approval for longer-range drone operations. Prime Air now operates from 11 locations across 10 U.S. metro areas, with more launches planned in cities including Chicago, Atlanta, Cleveland and Boise.

Faster Deliveries, Wider Selection

Prime Air drones can carry packages weighing up to 5 pounds and deliver them in as little as 30 minutes. Most orders currently arrive in about an hour.

Amazon says millions of products, including groceries, electronics, cosmetics and medications, are eligible for drone delivery.

The expansion is part of Amazon’s broader push toward faster delivery, as the company competes with other drone operators. Alphabet’s Wing has already surpassed 1 million commercial drone deliveries, while Zipline has completed 2 million deliveries across four continents.

Safety Challenges Remain

Despite the expansion, Prime Air continues to face safety and operational challenges. Two Amazon drones collided with a crane in Arizona last year, while other incidents have involved an internet cable in Texas and a drone crash in the UK.

Amazon says its drones use onboard cameras and sensors to detect obstacles and navigate safely, with no live camera feed monitored by people.

U.S. export controls restrict Nvidia’s most advanced AI chips from being shipped to China, but Chinese companies may still access their computing power through data centers in other countries.

The issue has gained attention as Chinese firms release increasingly capable AI models. In July, White House official Michael Kratsios accused Moonshot AI of using Nvidia GB300 chips through a facility in Thailand after the launch of its Kimi K3 model.

Follow THE FUTURE on LinkedIn, Facebook, Instagram, X and Telegram

How Chinese Firms Access Chips Abroad

Current U.S. rules primarily restrict the physical shipment and ownership of advanced AI chips, rather than remote access to computing power hosted overseas.

Cassia King of the Institute for AI Policy and Strategy told CNBC that Moonshot’s reported access through Thailand could be legal if the company did not own the physical hardware.

Chinese firms including ByteDance, Alibaba and Tencent have reportedly accessed Nvidia-powered infrastructure remotely through Thailand, Malaysia and Japan. ByteDance was also reportedly working with Singapore-based cloud provider Aolani to access Nvidia-powered computing in Malaysia.

Southeast Asia’s Growing Data Center Market

The development comes as data center construction accelerates across Southeast Asia. JLL estimates that global data center capacity could nearly double to 200GW by 2030, while DC Byte has identified 31 planned data centers of at least 100MW across Malaysia, Indonesia and Thailand.

That expanding infrastructure could give Chinese companies greater access to advanced computing without moving restricted chips into China.

Proposed Law Could Close The Gap

The proposed Remote Access Security Act would extend U.S. export controls to remote cloud access to critical hardware and software. It passed the House in January but still needs Senate approval.

Michelle Nie of the Center for a New American Security said the loophole undermines the goal of restricting China’s access to advanced U.S. chips.

However, legislation alone would not immediately solve the problem. Regulators would still need to determine which computing resources are restricted, who can access them and how cloud providers should verify customers.

According to King, creating rules could be done quickly with White House support, but making them effective and enforceable would be the bigger challenge.

Relativity Networks has raised $22 million as it develops hollow-core fiber technology designed to transmit data faster than conventional fiber and potentially expand where large AI data centers can be built.

The funding, announced Tuesday, was raised through SAFE notes from Rhapsody Venture Partners, Bell Ventures and Faster Than Glass, among others. The company also secured a $40 million follow-on order from an unnamed leading hyperscaler.

Follow THE FUTURE on LinkedIn, Facebook, Instagram, X and Telegram

Faster Fiber, Lower Latency

Relativity Networks uses hollow-core fiber, a technology that sends light through a hollow chamber rather than traditional glass fiber. The company says this can make data transmission around 30% faster.

According to CEO Jason Eisenholz, a signal takes roughly five microseconds to travel one kilometre through conventional fiber. Hollow-core fiber can reduce that to about 3.5 microseconds.

That difference becomes increasingly important as AI systems grow. Instead of keeping GPUs within a single data center, operators are increasingly spreading computing capacity across multiple buildings or campuses because of power and space constraints.

Extending The Reach Of AI Data Centers

Relativity believes its technology could help connect separate data centers and allow them to function as a single, synchronized system.

Reducing latency by 30% could effectively allow compute infrastructure to span greater distances before network delays become a major limitation.

“The first era of AI optimized for compute,” Eisenholz said. “The second era optimized the networking inside the data center. The third era that we see coming is optimizing the geography.”

Sen. Bernie Sanders has proposed legislation that would prevent the federal government from withholding Social Security benefits from older Americans and people with disabilities to repay defaulted federal student loans.

Sanders announced the Stop Social Security Garnishment Act on Monday. The bill is backed by Democratic Sens. Elizabeth Warren and Ed Markey and is expected to be formally introduced when the Senate returns next month.

Follow THE FUTURE on LinkedIn, Facebook, Instagram, X and Telegram

Millions Of Borrowers Are In Default

Nearly 9.5 million federal student loan borrowers were in default as of March, according to an Associated Press analysis of federal data. Sanders said nearly one in four borrowers cannot repay their loans and could face wage or Social Security garnishment.

Around 9.6 million borrowers aged 50 and older hold nearly $457 billion in outstanding student debt, according to Education Department data.

Collections Remain Paused

The proposal comes as the Trump administration has paused involuntary collections from borrowers in default.

In June 2025, the administration said it would not reduce Social Security benefits for affected borrowers, reversing an earlier plan to resume collections after pandemic-era protections ended. In January, the Education Department also announced a delay in wage garnishment and other involuntary collections while new repayment options were being implemented.

Borrowers in default have meanwhile been given more time to rehabilitate their loans and return to repayment.

What Sanders’ Bill Would Change

If passed, the legislation would prohibit the government from garnishing Social Security retirement and disability benefits to repay federal student loans. The measure would also protect older borrowers from forced collections that could affect their ability to pay for healthcare, medicine and other basic needs.

“In the richest country in the history of the world, no senior should have their Social Security payments taken away from them to pay back student debt,” Sanders said.

The proposal would not erase student debt, but would prevent Social Security benefits from being used to collect it.

Dcoop Reportedly Leads €470 Million Bid

Shares of Spanish olive oil producer Deoleo rose more than 15% on Wednesday as several companies compete to acquire the world’s largest olive oil bottler and marketer.

Spanish agri-food cooperative Dcoop has reportedly offered €470 million ($545 million), putting it ahead of Italian, French and Australian rivals. According to El Economista, the deal has not yet been finalized but is in its final stages, with a potential closing in September.

Follow THE FUTURE on LinkedIn, Facebook, Instagram, X and Telegram

If completed, the acquisition would strengthen Spain’s position in the global olive oil industry, creating a major producer with brands including Bertolli and Carbonell and an estimated 15% share of Spain’s domestic consumption.

Several Buyers Are In The Race

Dcoop is competing with Italy’s Coricelli, Bonifiche Ferraresi and Newlat Food, France’s Lesieur, owned by Avril, and Australia’s Cobram Estate Olive.

Deoleo shares were last up 15.4%, reaching a new 52-week high and heading for their strongest session since March 2022.

Olive Oil Market Stabilises

Spain, Italy and Greece remain among the world’s leading olive oil producers, while climate change, water shortages and pests have contributed to major price swings in recent years.

Deoleo recently told CNBC that the period of unprecedented volatility in the market has begun to give way to more stable conditions.

Humanoid Robot Maker Raises $905 Million In IPO

Shares of Chinese humanoid robot maker Unitree Robotics surged more than 460% in their Shanghai debut on Wednesday, after briefly jumping nearly 630%. The stock closed at 845 yuan.

The Hangzhou-based company raised about 6.1 billion yuan ($905 million) through its IPO, according to its prospectus.

Follow THE FUTURE on LinkedIn, Facebook, Instagram, X and Telegram

From Backflips To Industrial Robots

Unitree has gained international attention for its humanoid robots capable of walking, manipulating objects and performing acrobatic movements. Its portfolio also includes four-legged robots designed for applications such as hazard detection.

Ahead of the listing, the company unveiled “Superman”, a humanoid robot it says can jump two metres from a standing position and reach speeds of up to 12.66 metres per second.

Chinese AI company DeepSeek invested about 140.8 million yuan in Unitree, while existing investors include technology giant Tencent, according to company filings.

China’s Humanoid Robot Market Expands

Unitree is the latest major technology company to debut on Shanghai’s STAR Market. Memory chipmaker CXMT saw its shares jump 466% on their first trading day last month.

Morgan Stanley raised its forecast for China’s humanoid robot shipments to 50,000 units in 2026, nearly twice its previous estimate. The bank expects the country’s humanoid robotics market to grow from $2 billion this year to $15 billion by 2030.

Broader commercial deployments are expected to accelerate in the second half of 2026, with full-size humanoid robots projected to account for around 30% of shipments this year and 70% by 2028.

Labor Market Pressure Is Concentrated In A Few Industries

AI is beginning to weigh on employment across major developed economies, with the strongest effects appearing in highly exposed industries and among entry-level workers, according to Goldman Sachs.

Since late 2022, sectors more vulnerable to AI automation have generally seen slower growth in job openings, particularly in Germany, Australia and the U.S. Employment in information and communication services has also weakened across most developed economies, although it remains near or above its long-term trend outside the U.S.

Follow THE FUTURE on LinkedIn, Facebook, Instagram, X and Telegram

Call Centers Show The Clearest Impact

Call centers, software publishing, management consulting and advertising have seen employment fall well below historical trends. Call-center employment is now 39% below trend in the U.S., 33% lower in Canada and 27% lower in Germany.

Goldman’s analysis of more than 800 occupations found that entry-level workers face the strongest AI-related pressure. A 10% increase in occupational exposure to AI was associated with a 0.1 percentage-point drag on annual employment growth in France, Canada and the U.S., while the effect for entry-level workers was larger.

Still, Goldman said the impact remains concentrated in a relatively narrow group of industries and workers.

AI Adoption Continues To Grow

Goldman’s analysis of 11 surveys found AI adoption rates of around 15% to 20% across major developed economies. France, the U.S., the Netherlands and the U.K. were among the leaders, while Italy, Japan and New Zealand had lower adoption rates.

Major emerging markets recorded adoption rates of roughly 10% to 15%.

Record-High Stocks Face Growing Risks

U.S. and European stocks are reaching record levels as investors pour money into artificial intelligence, but economists at the European Central Bank warn that the current rally could eventually give way to a sharp correction.

In a Monday blog post, ECB economists said historical examples of major technological shifts suggest that current stock valuations are likely to fall at some point.

Follow THE FUTURE on LinkedIn, Facebook, Instagram, X and Telegram

One possible scenario is that excessive optimism pushes AI-related stocks above their fundamental value before investor confidence fades. Even if today’s valuations accurately reflect AI’s potential to transform the economy and increase corporate profits, the economists said a correction could still follow.

AI Boom Echoes Earlier Technology Waves

The ECB analysis compares the current AI investment cycle with past periods of rapid technological change, including the 19th-century railway boom, the expansion of electricity and radio in the 1920s, and the rise of the internet in the 1990s.

As new technologies become more widely adopted, uncertainty can spread across the broader economy. A major setback in the technology could then increase investors’ risk concerns and put downward pressure on stock prices, even if corporate profits remain strong.

According to the economists, these cycles typically involve a boom followed by a correction, potentially followed by another period of growth. However, they stressed that the timing of such a downturn cannot be predicted in advance.

Europe Could Be Particularly Exposed

European retail investors could face significant losses because global index and pension funds have substantial exposure to the so-called “Magnificent 7” U.S. technology companies.

A severe market correction could also create broader financial risks through investment funds and potentially affect euro-area stability. The ECB economists warned that policymakers may have less room than during the dot-com crash to respond, with fewer options to cut interest rates or use fiscal measures to cushion the impact.