&w=3840&q=100)
Gmail's Gemini-powered summaries may expose users to security risks: Report
Indian Express cites Marco Figueroa, GenAI Bug Bounty Programs Manager at Mozilla, stating that a security researcher uncovered a prompt injection flaw in Google Gemini for Workspace, which let attackers 'hide malicious instructions inside an email' that triggered when users clicked the 'Summarise this email' button in Gmail.
Attack through Gemini: How does this work
As per the report, hackers found a way to hide secret instructions in emails that trick Google's Gemini AI. They did this by placing hidden text at the end of the email using HTML and CSS, making the font size zero and the colour white so it could not be seen.
Because these emails do not contain attachments, they can easily pass through Google's spam filters and reach users' inboxes. When someone opens the email and clicks 'Summarise this email' using Gemini, the AI follows the hidden commands without knowing they are harmful.
Mozilla's Marco Figueroa explained how such prompt injections can be detected with:
Gemini can be updated to ignore or remove hidden text in emails.
Google can use a post-processing filter to scan Gemini's output for: Urgent messages, Phone numbers, Suspicious links.
These flagged elements can then be reviewed for potential threats.
Google has reportedly issued a statement to BleepingComputer, stating, 'We are constantly hardening our already robust defenses through red-teaming exercises that train our models to defend against these types of adversarial attack."
The company representative clarified to BleepingComputer that some of the mitigations are in the process of being implemented or are about to be deployed.
The report further states that Google has seen no evidence of incidents manipulating Gemini in the way demonstrated in Figueroa's report.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
&w=3840&q=100)
Business Standard
37 minutes ago
- Business Standard
CCI widens antitrust probe against Google in online ad tech segment
Search giant Google's run-in with India's competition watchdog appears to be intensifying. The Competition Commission of India (CCI) has expanded its investigation into the company's alleged anti-competitive practices in online advertising, based on a complaint filed by the Alliance of Digital India Foundation (ADIF). In an order passed on 1 August, the CCI said it would club the present matter with the ongoing investigation in similar cases. 'The DG is directed to investigate the various alleged practices of Google in online advertising services and/or AdTech intermediation services and submit a consolidated investigation report in the matter accordingly,' the order stated. Google, in a statement, said: 'We are reviewing the CCI's orders. We welcome the CCI's decision to dismiss one part of the complaint. We remain confident that our ongoing work with the CCI will affirm that Google's advertising practices have consistently benefited advertisers, publishers, and users, and are fully compliant with competition law.' The complaint outlined Google's dominant position and alleged abusive conduct in both the online search advertisement and online display advertisement markets. ADIF contends that Google's control over major online platforms, coupled with the fact that it derives 97 per cent of its revenue from advertising, has led to practices that 'stifle competition and adversely affect Indian businesses.' The CCI has been investigating Google since 2021, following cases filed by the Digital News Publishers Association. Two additional cases were subsequently filed. Google's tussles with Indian authorities have increased in recent years. In 2022, the CCI fined Google ₹936 crore for abusing its dominant position through its Play Store policy. The company was also fined ₹1,337.76 crore for its conduct in the Android mobile device ecosystem.
Mint
an hour ago
- Mint
RenAIssance of PR: CEOs must lead the narrative or be silenced by AI
One falsehood about AI (artificial intelligence) has been deliberately injected in its very name: that it is 'intelligent," even sentient. This is not true. It learns as it is taught. It speaks like a trained parrot. It is a case of reaping what you sow. I was a panelist at more than one CXO roundtable alongside the Cannes LIONS festival this year. We discussed messaging, brand risk, and the redefinition of news and views in the AI era. A clear takeaway was that CEOs need to take charge of PR and corporate reputation as core to business. Influence has gone beyond the human sphere to being computational. In a world flooded with synthetic content, algorithmic filters, and AI-generated headlines, the question is no longer whether your story will be told, but whether it will be surfaced, summarized, supercharged or silenced by machines. On paper, PR as an industry barely registers as a rounding off error in the global economy. But to dismiss it based on its size is to misunderstand its power. PR doesn't own platforms. It shapes perception. It doesn't control algorithms. It informs the inputs that train them. It may not scale like tech, but it steers what people believe about companies, crises, leaders, and entire industries. In the AI age, that influence is magnified. PR is increasingly embedded in the feedback loop of machines. Its narratives shape what language models surface, summarize, and reinforce. The most valuable compound in this era is 'fame with credibility'. PR is the machinery behind it. AI hasn't killed PR. In fact, it has made it more valuable than ever. What was once a media outreach function is now a boardroom lever for trust, visibility, and long-term value. We're not witnessing PR's decline. We're watching its reinvention. I call it a RenAIssance because it resets how businesses influence, communicate, and lead in a machine-shaped world. AI makes storytelling easier for all. Startups, NGOs, and challenger brands can wield the same communication tools as corporations paying multimillion-dollar retainers. There is far more noise and misinformation. PR can no longer be about gatekeeping. From the era of Edward Bernays to the era of ChatGPT, PR has evolved with every communication wave. The mission remains to earn trust, shape perception, and secure strategic relevance. For CEOs, it is a business-critical imperative, not just a media shift. PR once managed public opinion. Now it also needs to influence machine perception. Press releases, leadership op-eds, and earned media don't just sway human audiences , they train AI. Credible coverage feeds the large language models behind Google's Search Generative Experience and ChatGPT. These tools don't just surface links but 'scrape and summarise' from trusted sources. This shift from SEO (Search Engine Optimization) to GEO (Generative Engine Optimization) means discoverability now depends on authority, not keywords. In this funnel, PR is the feed. Ignore it, and your brand vanishes from the machine conversation. Marketing is stretched, stressed, and supercharged by AI. Campaigns go live in minutes. Messaging is automated. Targeting is machine-tuned. In this flood of machine-made content, authenticity is the buoy. PR brings the human voice back into the conversation. It needs to be trusted both by humans and machines. Done well, PR sharpens a i CEO's voice, strengthens stakeholder confidence, and enables long-term brand memory. In an attention-starved world, earned media, authentic storytelling, and editorial nuance are core assets, not optional extras. These stories get archived, cited, and resurface when it matters most. AI enables real-time sentiment tracking, narrative forecasting, and algorithmic reputation modelling. It allows the discourse to be shaped proactively. AI will automate drafting, data-crunching, and personalisation, but it's still human insight that delivers meaning. The messaging tone, ethical judgment, and necessary media relationships become the true differentiators. This is where PR becomes strategy, not support. In an age of deepfakes and eroding trust, PR must reinforce what's real. CEOs, pay attention — AI won't replace PR, but it will expose which organisations never took it seriously. If PR once seemed downstream of performance, today it's upstream. The future of PR isn't about faster press releases or clever headlines. It's about creating a brand story that humans believe in and machines amplify. Welcome to the RenAIssance. Shubhranshu Singh is a marketer and columnist. He was honoured as one of the 50 most influential global CMOs for 2025 by Forbes and serves on the board of the Effie LIONS foundation. The views expressed are personal.
Time of India
an hour ago
- Time of India
Google preserves popular goo.gl URLs, retires the rest
google In an important update to its link -shortening service, Google has announced it will preserve "actively used" links, while all other links created with the service will stop functioning on August 1st. The decision provides a reprieve for many users who were bracing for a complete shutdown of the legacy service. The company, which had previously announced the deprecation of clarified its new policy in a blog post aimed at developers and users. According to the post, links that have received a significant number of clicks or are embedded in frequently accessed content will be identified as "actively used" and will continue to work indefinitely. This move aims to prevent widespread link rot on the internet, a concern that had been raised by webmasters and digital archivists. For links that do not meet the "actively used" criteria, the redirection service will cease to function, resulting in a 404 error for users who click on them. Google encouraged creators of these links to update their content with a new redirection service. 'Nine months ago, we redirected URLs that showed no activity in late 2024 to a message specifying that the link would be deactivated in August, and these are the only links targeted to be deactivated. If you get a message that states, 'This link will no longer work in the near future', the link won't work after August 25 and we recommend transitioning to another URL shortener if you haven't already. All other links will be preserved and will continue to function as normal. To check if your link will be retained, visit the link today. If your link redirects you without a message, it will continue to work,' Google said in a blog post. The move underscores the challenge of managing a vast and ever-changing internet infrastructure. While the transition may still require some effort from users and developers, Google's decision to maintain active links shows a commitment to preserving the web's functionality and historical content.
