Tech
Study Finds AI Models Get Basic Math Wrong Around 40 Percent of the Time
Artificial intelligence (AI) tools are increasingly used for everyday calculations, but a new study suggests users should approach their answers with caution. Researchers from the Omni Research on Calculation in AI (ORCA) found that when tested on 500 real-world math prompts, AI models had roughly a 40 percent chance of producing an incorrect result.
The study evaluated five widely used AI systems in October 2025: ChatGPT-5 (OpenAI), Gemini 2.5 Flash (Google), Claude 4.5 Sonnet (Anthropic), DeepSeek V3.2 (DeepSeek AI), and Grok-4 (xAI). None of the models scored above 63 percent overall, with Gemini leading at 63 percent, Grok close behind at 62.8 percent, and DeepSeek at 52 percent. ChatGPT-5 scored 49.4 percent, while Claude trailed at 45.2 percent. The average accuracy across all five models was 54.5 percent.
“Although the exact rankings might shift if we repeated the benchmark today, the broader conclusion would likely remain the same: numerical reliability remains a weak spot across current AI models,” said Dawid Siuda, co-author of the ORCA Benchmark.
Performance varied across categories. AI models performed best in basic math and conversions, with Gemini achieving 83 percent accuracy and Grok 76.9 percent. ChatGPT-5 scored 66.7 percent in the same category, giving a combined average of 72.1 percent—the highest across the seven tested categories. Physics proved the most challenging, with overall accuracy dropping to 35.8 percent. Grok led this category at 43.8 percent, while Claude scored just 26.6 percent.
Some AI systems struggled more than others in specific fields. DeepSeek recorded only 10.6 percent accuracy in biology and chemistry, meaning it failed nearly nine out of ten questions. In finance and economics, Gemini and Grok reached 76.7 percent, while the other three models scored below 50 percent.
The study also categorized the types of mistakes AI makes. “Sloppy math” errors, including miscalculations or rounding issues, accounted for 68 percent of mistakes. Faulty logic errors represented 26 percent, reflecting incorrect formulas or assumptions. Misreading instructions accounted for 5 percent, while some AI simply refused to answer. Siuda noted that multi-step calculations with rounding were particularly prone to error.
The research highlights the importance of verifying AI-generated calculations. “If the task is critical, use calculators or proven sources, or at least double-check with another AI,” Siuda advised.
All 500 prompts used in the study had one correct answer and were designed to reflect everyday math tasks, including statistics, finance, physics, and basic arithmetic. The findings indicate that while AI can assist with calculations, it remains unreliable for precise numerical work and users should remain cautious when relying on these tools.
Tech
Cyberattacks Intensify as Iran Conflict Spills Into Digital Domain
State-linked and hacktivist groups have claimed a series of cyberattacks against the United States and Israel since the war with Iran began, marking a significant escalation in the digital dimension of the conflict.
One of the most notable incidents involved Stryker, which confirmed on March 11 that a cyberattack had disrupted its global network. According to reports, employees encountered the logo of Handala, an إيران-linked hacking group, on login pages across the company’s systems. The breach reportedly targeted the firm’s Microsoft-based infrastructure, though the full extent of the disruption remains unclear.
Handala has claimed responsibility for the attack, stating it exploited cloud management systems to remotely wipe large numbers of devices worldwide. The group said the operation was carried out in retaliation for a missile strike in Iran. Independent verification of these claims is still pending.
Cybersecurity analysts say the attack is part of a broader campaign by groups linked to Iran’s security apparatus. According to findings from CloudSek, organisations associated with the Islamic Revolutionary Guard Corps have targeted US critical infrastructure. These include CyberAv3ngers, APT33 and APT55, which are accused of attempting to infiltrate industrial systems such as power grids and water facilities.
Experts say some of these groups use simple methods, including default passwords, to access systems, while others deploy malware aimed at disrupting operations or gathering intelligence. Additional networks linked to Iran’s Ministry of Intelligence have also been active, targeting telecommunications, energy companies and government organisations.
At the same time, the United States and Israel are conducting their own cyber operations. General Dan Caine said US Cyber Command played a key role early in the conflict, disrupting Iranian communications and sensor networks. Defence Secretary Pete Hegseth confirmed that artificial intelligence and cyber tools are being used alongside conventional military operations.
Israeli intelligence has also reportedly relied on hacked data to support military planning, highlighting the growing role of cyber capabilities in modern warfare.
Hacktivist activity has surged as well. More than 60 groups formed a loose coalition known as the Cyber Islamic Resistance, coordinating attacks through online platforms. These groups have claimed hundreds of operations, including attempts to disrupt Israeli infrastructure and private sector systems. Analysts warn that such actors are often less restrained and may pose risks to civilian networks.
The conflict has also drawn in groups from outside the region, including actors based in Iraq, Russia and other parts of the Middle East. Some have targeted government websites and transport infrastructure, while pro-Israeli groups have carried out retaliatory attacks against Iranian entities.
Security experts say the growing scale and coordination of cyber operations reflect a shift in how modern conflicts are fought, with digital attacks now running parallel to military action on the ground.
Tech
Study Finds Hormone-Disrupting Chemicals in Popular Headphones Sold Across Europe
Tech
China Approves First Commercial Brain Implant as Neuralink Plans Mass Production
China has granted regulatory approval for the world’s first brain implant intended for commercial use, offering new hope for people with paralysis to regain hand movement. The device, developed by Neuracle Medical Technology, employs a brain-computer interface (BCI) that translates brain signals into physical actions.
BCIs link the nervous system to external devices, allowing users to control technology or prosthetics purely with thought. Neuracle’s system targets individuals whose paralysis stems from severe spinal cord injuries in the neck, which block signals from the brain from reaching the arms and hands.
The implant detects neural signals associated with the intent to move the hand. These signals are interpreted by software and transmitted to a robotic glove worn by the patient. The glove, powered by air-driven mechanisms, enables the hand to open and close, allowing users to grasp objects, according to CGTN.
Eligibility is limited to adults aged 18 to 60 who have experienced paralysis for at least one year and whose condition has remained stable for six months. The device is intended for patients unable to grip objects with their hands but who retain some movement in their upper arms.
China has been ramping up its investment in BCI technology, naming it a national strategic priority and highlighting it as a potential driver of future economic growth. Recent achievements include a successful implant by Shanghai NeuroXess, which allowed a 28-year-old man paralyzed for eight years to control digital devices with his thoughts within five days of receiving the implant.
The Neuracle approval comes as the race to commercialize BCIs intensifies worldwide. US entrepreneur Elon Musk, whose company Neuralink began human trials in 2024, recently announced plans to begin “high-volume production” of Neuralink devices in 2026.
As of September 2025, 12 participants with severe paralysis had received Neuralink implants, enabling them to operate digital and physical tools with thought alone. Musk’s announcement signals the company’s intent to scale access to BCIs beyond initial trials, positioning both China and the US at the forefront of this emerging field.
The development highlights a significant milestone in neurotechnology, potentially transforming the lives of millions living with paralysis. By translating intent into motion, these devices promise to restore independence to those previously constrained by spinal injuries, while also underscoring the global momentum toward commercial BCI applications.
With China now officially approving a commercial implant and Neuralink preparing for mass production, the coming years could see rapid adoption of technologies that bridge the human mind and machine.
-
Entertainment2 years agoMeta Acquires Tilda Swinton VR Doc ‘Impulse: Playing With Reality’
-
Business2 years agoSaudi Arabia’s Model for Sustainable Aviation Practices
-
Business2 years agoRecent Developments in Small Business Taxes
-
Home Improvement1 year agoEffective Drain Cleaning: A Key to a Healthy Plumbing System
-
Politics2 years agoWho was Ebrahim Raisi and his status in Iranian Politics?
-
Business2 years agoCarrectly: Revolutionizing Car Care in Chicago
-
Sports2 years agoKeely Hodgkinson Wins Britain’s First Athletics Gold at Paris Olympics in 800m
-
Business2 years agoSaudi Arabia: Foreign Direct Investment Rises by 5.6% in Q1
