The vulnerability was discovered by the Cato CTRL research group of cybersecurity company Cato Networks and originates in common AI tools, including Google's Gemini, Microsoft's Copilot, and Perplexity's Comet.

The research demonstrated primary attack scenarios in which attackers direct AI assistants to display fake phone numbers and links to users when they request customer service contact information for various organizations. The scenarios included extracting sensitive user data and sending it to malicious sources without the user's knowledge, stealing login credentials, displaying false information, and creating fake narratives that could influence the user and lead to wrong decisions.

The technique used by attackers is called HashJack. All they need to do is add a prompt – malicious instructions – to a legitimate website address and distribute it. Once a user loads the website address with the malicious addition in the browser, the instructions "communicate" with smart AI assistants, such as Google's Gemini or Microsoft's Copilot, and trigger attack scenarios.

According to Cato Networks, traditional defense systems do not detect the attack because they operate through prompts (instructions) embedded in the website address after the hashtag symbol # in a process that does not leave the browser's work.

The attack exploits users' trust in legitimate websites by using link addresses that appear legitimate. The user has no reason to suspect at any stage of the process, unlike phishing sites that look suspicious. This way, any legitimate site could become an attack tool – with attackers not even needing to breach the site itself. They exploit how AI browsers interpret instructions after the hashtag symbol. This effectively creates a new subcategory of cyber threats in the AI world.

According to the company's statement, the companies whose tools the vulnerabilities were identified in were informed well in advance of the problems so they could address them before users were exposed to threats (a practice known in the cyber field as "white hat hacker" hacking). According to Cato's data, a fix was applied in the Copilot for the Edge browser on October 27, 2025. In the Comet browser, a fix was reported to have been applied on November 18, 2025. In the Gemini for Chrome browser, as of November 25, 2025, the problem has not yet been resolved.

The post Israeli tech firm finds AI vulnerability – Gemini susceptible appeared first on www.israelhayom.com.

This significant upswing demonstrates a marked shift in attitudes toward Alphabet after the 2022 introduction of ChatGPT led certain investors to worry the firm had lost its AI advantage, despite having invented much of the foundational technology for generative AI. Alphabet recovered its momentum this year by making its cloud division a key contributor to growth, which attracted an investment from Warren Buffett's Berkshire Hathaway, and its new Gemini 3 model is garnering excellent initial reviews.

The milestone, however, could intensify anxieties among some executives who caution that rising valuations have detached market trends from business reality, prompting concerns over an economic bubble reminiscent of the 1990s dot-com era. These bubble fears are also amplified by circular transactions involving OpenAI and Nvidia, two prominent firms in the AI sector. Analysts, nevertheless, maintain that Google is strongly positioned in the AI competition because of its ample cash reserves, its own in-house chip production offering an alternative to Nvidia's expensive processors, and its massive internet search operation, which is already experiencing benefits from AI incorporation.

In October, the moment of truth arrived for the AI tech titans as investors worldwide held their breath after Wall Street's close when Microsoft, Alphabet (Google), and Meta (Facebook) – collectively worth nearly 10 trillion dollars – began releasing their quarterly earnings reports in another attempt to beat forecasts.

Google (through its parent company Alphabet) beat forecasts by 26% ($2.87 per share compared to the $2.27 expected) with revenues of 100 billion dollars. Microsoft also beat forecasts by 14% ($3.72 per share compared to the $3.6 expected). Meta on the other hand had to pay 15 billions in a one-time payment, which resulted in a miss of 84% in the earnings.

Alphabet reported more than $100 Billion of revenue in a quarter for the first time ever. Google parent company Alphabet also reported strong third-quarter results, with revenue climbing 16% year-over-year. Google Search & Other grew 15%, while YouTube Ads increased 15% during the period. Google Cloud posted the strongest growth at 34%, though Google Network saw a 3% decline. The company's EBIT rose 9.5% with a margin of 31%, and earnings per share jumped 35%.

The head of Google's parent firm Alphabet, Sundar Pichai, has recently cautioned the BBC that every company would be impacted if the AI bubble were to burst. Speaking exclusively to BBC News, Pichai acknowledged that while AI investment growth is an "extraordinary moment," there is some "irrationality" in the current boom. Amid fears in Silicon Valley of soaring valuations, Pichai warned, "I think no company is going to be immune, including us."

His wide-ranging interview at Google's California headquarters, reported by the BBC, touched on energy needs, climate targets, UK investment, and AI's effect on jobs, coming as market scrutiny intensifies. Alphabet's shares have doubled in seven months to $3.5 trillion as market confidence in its ability to counter OpenAI grows. A specific focus remains on Alphabet's competition with Nvidia, run by Jensen Huang, which recently hit a $5 trillion valuation, even as analysts question the $1.4 trillion in deals surrounding OpenAI.

Pichai's remarks about "irrationality" echoed Alan Greenspan's 1996 "irrational exuberance" warning during the dotcom boom. Pichai told the BBC the industry can "overshoot" in such cycles. "We can look back at the internet right now. There was clearly a lot of excess investment, but none of us would question whether the internet was profound," he said. "I expect AI to be the same. So I think it's both rational and there are elements of irrationality through a moment like this." His comments follow a similar warning from JP Morgan's Jamie Dimon, who told the BBC some AI investments would "probably be lost." Pichai argued Google's "full stack" model provides a better position to handle turbulence.

The post Alphabet's market cap approaches $4 trillion amid AI rally appeared first on www.israelhayom.com.

His wide-ranging interview at Google's California headquarters, reported by the BBC, touched on energy needs, climate targets, UK investment, and AI's effect on jobs, coming as market scrutiny intensifies. Alphabet's shares have doubled in seven months to $3.5 trillion as market confidence in its ability to counter OpenAI grows. A specific focus remains on Alphabet's competition with Nvidia, run by Jensen Huang, which recently hit a $5 trillion valuation, even as analysts question the $1.4 trillion in deals surrounding OpenAI.

Pichai's remarks about "irrationality" echoed Alan Greenspan's 1996 "irrational exuberance" warning during the dotcom boom. Pichai told the BBC the industry can "overshoot" in such cycles. "We can look back at the internet right now. There was clearly a lot of excess investment, but none of us would question whether the internet was profound," he said. "I expect AI to be the same. So I think it's both rational and there are elements of irrationality through a moment like this." His comments follow a similar warning from JP Morgan's Jamie Dimon, who told the BBC some AI investments would "probably be lost." Pichai argued Google's "full stack" model provides a better position to handle turbulence.

The BBC noted the tech giant is also expanding its UK footprint, having announced a £5 billion investment in UK AI infrastructure and research in September. Pichai confirmed Alphabet will develop "state-of-the-art" research via its London-based DeepMind unit. He stated for the first time that Google would "over time" "train our models" in the UK, a move ministers believe will secure the UK as an AI "superpower" after the US and China. "We are committed to investing in the UK in a pretty significant way," Pichai said.

However, the Alphabet chief also highlighted the "immense" energy requirements of AI, which the International Energy Agency stated consumed 1.5% of global electricity last year. Pichai, according to the BBC, said action is needed, including in the UK, to scale up energy infrastructure. "You don't want to constrain an economy based on energy, and I think that will have consequences," he warned. He conceded these intensive energy needs mean slippage on the company's 2030 net-zero climate target, stating, "The rate at which we were hoping to make progress will be impacted."

AI's impact on employment will be significant, Pichai told the BBC, labeling it "the most profound technology" humanity has ever addressed. "We will have to work through societal disruptions," he stated, noting it will simultaneously "create new opportunities." He explained, "It will evolve and transition certain jobs, and people will need to adapt," emphasizing that those who adjust to AI "will do better," adding, "It doesn't matter whether you want to be a teacher [or] a doctor. All those professions will be around, but the people who will do well in each of those professions are people who learn how to use these tools."

The post Google chief concedes 'irrationality' in AI boom, warns of impact if bubble bursts appeared first on www.israelhayom.com.

The US Justice Department and several states have asked a federal court to compel Google to sell Chrome, its widely-used web browser, in a move that could reshape the $2 trillion company's future and transform internet competition, according to reporting from The New York Times. The request, submitted Wednesday, follows Judge Amit P. Mehta's August ruling in the US District Court for the District of Columbia that found Google had illegally maintained a search monopoly.

The government's proposal, marking the most significant tech antitrust remedy since the 2000 Microsoft case, also gives Google an ultimatum: either sell Android, its smartphone operating system or eliminate mandatory Google services on Android devices. Chrome commands roughly 67% of the global browser market, while Android dominates mobile software with a 71% market share, according to Statcounter. "The playing field is not level because of Google's conduct, and Google's quality reflects the ill-gotten gains of an advantage illegally acquired," government officials stated in their filing.

Kent Walker, Google's president of global affairs, strongly contested the government's stance. "DOJ's wildly overbroad proposal goes miles beyond the court's decision," he wrote in a blog post, arguing it would disrupt Google products that consumers find valuable. The company is expected to submit its own remedial proposals by Dec. 20, with Judge Mehta scheduled to hear arguments this spring and issue a ruling by summer's end.

The case stems from Google's practice of paying companies like Apple, Mozilla, and Samsung to be the default search engine on their devices, with payments reaching $26.3 billion in 2021, according to trial evidence. While Google maintains users choose its service for superior performance, the Justice Department argued these agreements cemented the company's market dominance by ensuring robust search traffic and data collection.

The government's filing also seeks to restrict Google's artificial intelligence ventures, requesting forced divestiture of stakes in AI companies that could compete in search, including its investment in Anthropic, maker of the Claude chatbot. Additionally, the proposal would require Google to allow publishers to opt out of having their content used for AI model training, setting up another significant battleground in the ongoing tech regulation debate.

The post End of Google's monopoly? appeared first on www.israelhayom.com.

Google's worldwide employee base, including its Israeli contingent, received a $400 charitable giving allocation on Monday, with the option to direct these funds to organizations from a company-approved list. The inclusion of UNRWA – the UN agency whose operations in Israel were recently suspended following evidence of its personnel's participation in the Oct. 7 Hamas attacks – has sparked considerable discussion within the company.

The charitable initiative allows employees to designate their $400 donation to any organization from an extensive list of over 1000 verified charities working across various sectors. The presence of UNRWA and its affiliated organizations on this list was quickly flagged by Israeli employees and pro-Israel staff members in an internal chat channel.

The scale of Google's charitable initiative is substantial, with allocations made to approximately 180,000 employees, representing a potential charitable impact of $72 million. The company has demonstrated strong support for Israeli emergency services since the outbreak of the Iron Swords War, actively encouraging global employee donations to organizations including Magen David Adom, United Hatzalah, and the ERAN crisis hotline.

The controversy emerges against the backdrop of unprecedented legislative action in the Knesset, which last month passed two significant laws restricting UNRWA's operations. The first legislation bars the agency from operating in Jerusalem's Shuafat refugee camp and terminates all its activities in east Jerusalem. The second law establishes a firewall between Israeli government officials and UNRWA personnel, effectively preventing the agency's staff from working in or accessing Gaza.

The legislative framework explicitly prohibits UNRWA from maintaining any presence, providing services, or conducting activities, whether directly or indirectly, within Israel's sovereign territory.

In response to inquiries, Google Israel stated: "Google employees can select from over 200,000 organizations available through the Benevity platform, a third-party provider responsible for validating eligibility for corporate donation programs. These organizations must regularly demonstrate their compliance with the program's eligibility criteria."

The post Google backs UNRWA donation despite Oct. 7 involvment appeared first on www.israelhayom.com.

In her inaugural earnings call as Alphabet's chief financial officer, Israeli-American Anat Ashkenazi reported robust third-quarter results while outlining a strategic vision balancing AI investments with fiscal discipline. The tech giant posted a 15% year-over-year revenue increase to $76.5 billion under her first quarterly review.

Ashkenazi's journey to leading finance at one of the world's largest tech companies reflects a notable trajectory through Israel's academic and banking sectors. After earning her bachelor's degree in accounting and economics from the Hebrew University of Jerusalem and an MBA from Tel Aviv University, she began her career at Bank Hapoalim, one of Israel's largest financial institutions, before joining Eli Lilly in 2001.

"As I look at the business, I see opportunities for further growth propelled by AI and the underlying momentum across the business," Ashkenazi told investors during Thursday's earnings call for Alphabet, the parent company of Google. "While we have a strong balance sheet to support these investments, we will be looking for efficiencies so that we can fund innovation in priority areas." Her debut quarter demonstrated impressive performance across key metrics, with operating income jumping 34% to $28.5 billion and earnings per share increasing 37% to $2.12. The transition from her predecessor Ruth Porat appears seamless, with Ashkenazi already implementing her vision for the company's financial future.

Drawing on financial expertise developed across multiple continents, Ashkenazi detailed significant infrastructure investments, with capital expenditure reaching $13 billion during the quarter. The company announced over $7 billion in planned data center investments, with nearly $6 billion allocated to US facilities. The cloud division showed particularly strong results under her first quarter, with revenue surging 35% to $11.4 billion and operating margins reaching 17%. YouTube achieved a milestone during her initial quarter, with combined advertising and subscription revenue surpassing $50 billion over the past four quarters.

Google's new CFO, Anat Ashkenazi, is cutting costs & boosting AI spending . She's looking to squeeze more savings out of the company while investing in AI innovation. What's next for Google's AI plans? #AI #Googlehttps://t.co/seWvxscNVq

— AI News Bulletin (@ai_bulletin_top) October 29, 2024

Alphabet CEO Sundar Pichai highlighted the other aspects of the report while providing a more refined outlook. "Q3 was another great quarter. The momentum across the company is extraordinary," said Pichai. "Our commitment to innovation as well as a long-term focus and investment in AI are paying off."

Looking ahead, Ashkenazi indicated that advertising revenue growth in Q4 will face tough comparisons due to strong performance in the second half of 2023, summing up the overall gains made in the reporting period. "In Q3, we also returned value to shareholders in the form of $15.3 billion in share repurchases and $2.5 billion in dividend payments. Overall, we returned a total of nearly $70 billion over the trailing 12 months to shareholders. As we look forward, we're working to balance our investments in AI and other growth areas with the cost discipline needed to fund those activities," Ashkenazi noted.

The post Google's Israeli-American CFO delivers strong results appeared first on www.israelhayom.com.

Cloud security startup Wiz is targeting $1 billion in annual recurring revenue (ARR) by 2025, according to co-founder Roy Reznik in an interview with CNBC. The company, which recently opened its first European office in London, is also considering an initial public offering (IPO) when market conditions are favorable.

"We've already broken a few records as a private company, and we believe we can also break a few more records as an independent public company as well," Reznik told CNBC at Wiz's new London office. The four-year-old firm has raised $19 billion in venture capital to date, including $1 billion secured this year at a $12 billion valuation.

Wiz, founded by four friends who served in Israel's intelligence Unit 8200, specializes in software that scans cloud environments for potential security risks. The company has gained significant traction in the rapidly growing cloud security market, with Reznik noting, "The cloud market is going to reach $1 trillion over the next few years."

Earlier this year, Wiz reportedly turned down a $23 billion acquisition offer from Google. CEO Assaf Rappaport stated at the time that the company was "flattered" but intended to remain independent and pursue an IPO.

Reznik, who serves as vice president of research and development, told CNBC that Wiz has received offers from "many people that want to get their hands on Wiz stock." However, he emphasized the company's commitment to independence and its IPO aspirations.

The startup has experienced rapid growth, reaching $100 million in ARR just 18 months after achieving $1 million. Wiz claimed this made it "the fastest software company to achieve this feat." Now, with $500 million in ARR, the company aims to double that figure by 2025.

Wiz's expansion into the UK market represents a significant growth opportunity, according to Reznik. "I think the talent here is amazing, and the ecosystem is amazing," he said. "We have always been very much involved in Europe – and specifically the UK – and I feel like it's a natural evolvement of Wiz to double down even more here in London and the UK."

Recent high-profile cyberattacks in the UK, including incidents affecting National Health Service hospitals and Transport for London, have increased interest in Wiz's offerings. The company already counts online retailer ASOS and digital bank Revolut among its UK customers.

Reznik, who recently relocated from Israel to the UK, sees tremendous potential in the British market. "We see a lot of interest here. Many of the largest banks and retailers, are Wiz customers. But we're also seeing a huge potential for growth," he told CNBC.

The post Israeli cloud security startup Wiz sheds light on IPO prospects appeared first on www.israelhayom.com.

Google has introduced Imagen 3, an advanced AI system that generates high-quality, photorealistic images up to 1024x1024 pixels from text prompts, outperforming similar tools like Midjourney, DALL-E 3, and Grok-2.

Imagen 3 uses a latent diffusion model trained on a large dataset of images, text, and annotations, with improved prompt understanding and safeguards to prevent offensive or illegal content generation.

It allows users to create, edit, and upscale images by providing detailed instructions and selecting specific areas for adjustments, with an interactive approach and user-friendly design. Some users were frustrated with its censorship mechanisms, as non-problematic prompts are also being blocked.

Google included its digital watermark SynthID for origin tracking to minimize harmful content. Imagen 3 is initially available only in the U.S. through the AI Test Kitchen service, ImageFX, and Vertex AI, with plans to expand to other countries.

Google may be taking a cautious approach with Imagen 3's rollout, possibly influenced by backlash against the Gemini model for producing historically inaccurate images, and trying to avoid controversy experienced with earlier image creation models.

Sources: ZDNet, PCMag, Android Central, Times of India, Deccan Herald, PetaPixel, Mobile Bulgaria, Chip Turkey, TechGear, Analytics Insight, RM Update, Times Now, PCM, Rosario3, Planetared, Notebookcheck

This article was written in collaboration with Generative AI news company Alchemiq.

The post Google introduces Imagen 3, an advanced AI image generator appeared first on www.israelhayom.com.

Google's AI company DeepMind developed an AI robotic system that can play table tennis at an intermediate amateur level, winning around 45% of matches and 46% of individual games against human players of varying skill levels. It excelled against beginners, won 55% against intermediate players, but lost to advanced players.

The robot combines an industrial robotic arm with custom AI software, with high-speed cameras to track the ball and respond effectively to human players' moves. It can execute various shots, read ball tilts, and respond to different ball speeds.

The robotic system was trained on a comprehensive dataset of table tennis ball states and practiced skills like forehand topspin and backhand targeting in a simulated environment. Its training involved reinforcement learning, imitation learning, and learning from real-world ball trajectories and human-vs-human gameplay data.

The robot adapted to different playing styles by tracking opponents' behaviors and adjusting its strategy through repeated training cycles involving reinforcement learning with real-world data.

Table tennis has been a standard for robotic research due to its demands on skills and planning. The achievement is viewed as a step towards achieving consistent human-level performance in table tennis with robots.

Researchers propose investigating advanced control algorithms and hardware optimizations to improve the robot's performance, particularly in handling fast balls, and making its play more unpredictable by allowing it to learn from human opponents' strategies.

The research represents progress towards creating robots that can perform tasks skillfully and safely in real environments. Playing against the AI robot was generally fun and engaging for most human players.

The robot arm is mounted on linear gantries for sideways movement and holds a 3D-printed paddle. The AI system has low-level skill controllers and a high-level strategic decision-maker, allowing it to execute table tennis techniques and adapt its strategy.

Table tennis is chosen by many robotics companies, including OMRON's "FORPHEUS" robot table tennis tutor, to train their systems due to its requirements.

The robot could potentially be used to help athletes train and create unexpected plays. Table tennis was chosen as a sport due to the complex physical and cognitive elements involved.

Sources: Technology Review, Live Science, TechRadar, Digit, The Next Web, Times of India, The Valley Post, Brytfmonline, Ex Bulletin, NY Breaking, The Times Hub, Singularity Hub, Techeblog, Faharas, Designboom, Tech Xplore, Cybernews, Interesting Engineering, NewsBytes.

This article was written in collaboration with Generative AI news company Alchemiq.

The post Google DeepMind's table tennis robot outperforms some human players appeared first on www.israelhayom.com.

Google announced the Google TV Streamer, a new premium 4K streaming device priced at $99.99, designed to compete with high-end streaming devices like Apple TV 4K and Roku Ultra, offering faster performance, improved features, and a redesigned interface.

The TV Streamer supports 4K at 60 frames per second, Dolby Vision, HDR10, HDR10+, Dolby Atmos, spatial audio, and has amenities like an Ethernet port, 32GB of built-in storage, and a faster processor (22% faster CPU, more RAM) than previous models. It also includes a built-in Thread border router for improved connectivity.

The device offers enhanced search capabilities with Gemini AI, providing curated recommendations, show summaries, reviews, and season breakdowns based on streaming subscriptions and preferences. It integrates with Google Photos and supports features like ambient mode and spatial audio.

The Google TV Streamer functions as an entertainment device and smart home hub, integrating with Google Home and Matter to control various smart home devices through voice commands and routines. It also serves as a control panel for viewing security cameras and receiving alerts from appliances.

The device has a wireless remote with improved grip, programmable buttons, and a function to search for lost remotes. It supports voice commands.

The Google TV Streamer is scheduled for release on September 24, with preorders starting earlier. Google plans to eventually phase out the Chromecast in favor of the TV Streamer,but will continue to support existing Chromecast devices with updates. Google may introduce a cheaper, HD-only version of the new set-top box in the future.

The Chromecast brand will be available while supplies last, but newer devices like the Chromecast with Google TV will continue receiving software and security updates.

While the TV Streamer supports a wide range of apps and over 800 free channels, its $100 price point may be a barrier for those who only stream in HD, and it lacks appeal to the gaming market.

Google has also unveiled the new Nest Learning Thermostat (4th Gen) with a new design inspired by the Pixel Watch, featuring a slimmer, more premium look, a larger screen, rotating metal dial, and made with recycled materials. It supports Matter, has a dynamic farsight feature showing weather and temperature, and comes with a redesigned Nest Temperature Sensor for better whole-home temperature tracking. It is available for pre-order at $279.99 in the U.S.

Sources: Mobile Time, Business Insider, Wired, Ars Technica, Forbes, USA Today, CNN, CNET, The Verge, Engadget, PCMag, TechCrunch, TechRadar, Mashable, Gizmodo, ZDNet.

This article was written in collaboration with Generative AI news company Alchemiq.

The post Google introduces a $99.99 Google TV Streamer, to replace Chromecast appeared first on www.israelhayom.com.