Super important, very critical: Google India head on tackling misinformation, deepfakes in AI era

New technologies like AI have opened massive opportunities but pose challenges like
deepfakes
,
Google India
's Country Manager and Vice President
Preeti Lobana
said asserting that fighting
misinformation
remains a top priority for the tech giant, which relies on strong policies, advanced
AI technology
, and human oversight to address these issues systematically.
The Asia Pacific region is particularly seeing higher degree of scams/frauds and misinformation has been a challenge, she said adding Google is stepping up its efforts to curb misleading and fake content.
Noting that Google last year had announced plans for launching 'Google Safety Engineering Centre' in India, Lobana termed the move "imminent".
"This (tackling misinformation) is super important for us, when you think about our mission, about information being universally accessible and organising it in a certain way, making sure that we are tackling misinformation in a very systematic manner is very, very critical. So...(it is about) having the right policies and guidelines, having the right technology, having the right human oversight to make sure that we are catching misinformation," she said.
With innovations like
SynthID
for
watermarking
and verifying AI-generated content, Google is also working with partners to strengthen
content authenticity
and trust.
"...we're introducing innovation like SynthID, so when any content is created using some of Google's AI tools, there is an invisible watermark, and it's pretty strong technology, because even if it is shared across multiple people or edited, it is detectable," she said.
Google has also introduced a SynthID verifier that allows users to upload content and detect whether or not a video is synthetic or AI generated.
Pointing out that fight against misinformation is always a work in progress she said Google is collaborating with other stakeholders in the ecosystem as well on provenance and authenticity.
" These are our efforts. The ecosystem needs to come together, but it is deeply important to us to make sure that we are combating that," she said.
Noting that Google had announced last year plans to launch Google Safety Engineering Centre in India, Lobana termed the move "imminent". On Tuesday, the company also unveiled Google Safety Charter for India's AI-led transformation.
As per Google, the charter is a blueprint for tackling the online world's new challenges collaboratively with the wider ecosystem. This includes keeping the end user safe from online frauds and scams; cybersecurity for government and enterprise infrastructure; and building AI responsibly.
Lobana said while AI, like other technologies in the past, had unlocked creative potential, it had fuelled a surge in misinformation and deepfakes.
"Therefore our effort is to make sure that whatever content is created using our AI, there are watermarks on that, and then (the idea is) enabling and sharing tools through which a wider section of users can upload some of this content to be able to identify it. But like I said, it is about working with a broader ecosystem as well, because multiple AIs are used to generate some of this content," she said.
Combating misinformation and deepfakes is a work in progress and an area of deep focus for not just Google, but others in the industry, Lobana noted.

Try Our AI Features

Explore what Daily8 AI can do for you:

Learn with AIFact CheckingText to SpeechAI Summary

Related Articles

Deccan Herald

an hour ago

• Deccan Herald

Samsung to launch Galaxy M36 5G next week in India

Thanks to deeper collaboration with Google and Samsung, the Galaxy M36 will support advanced versions of Gemini AI features. It will be priced under Rs 20,000 in India.

First Post

an hour ago

• First Post

16 billion passwords compromised, says report; have you changed yours?

A massive breach has exposed over 16 billion usernames and passwords from platforms like Google, Apple, Facebook, and more. The leak raises serious cybersecurity concerns, prompting urgent calls for stronger passwords, two-factor authentication, and regular dark web exposure checks. read more A staggering 16 billion usernames and passwords have been exposed in what experts are calling the largest-ever database of stolen credentials. The trove of compromised data includes login details from major platforms such as Apple, Google, Facebook, Telegram, GitHub and even government services, raising alarms over the global state of digital security. Cybersecurity researchers say the breach stems from a collection of 30 massive datasets, each holding tens of millions to over 3.5 billion records. The information, mostly acquired through infostealing malware, appears to be freshly leaked, with nearly all of the datasets previously unreported except for one earlier disclosure of 184 million passwords by researcher Jeremiah Fowler, according to a new investigation by Cybernews. STORY CONTINUES BELOW THIS AD 'Most of these credentials are structured as URLs followed by usernames and passwords, and they cover virtually every type of online service imaginable,' said Vilius Petkauskas, a Cybernews analyst who has been investigating the leak since the beginning of the year. The scale of this breach surpasses previous incidents, including last year's so-called 'Mother of All Breaches' which exposed 26 billion records. While it's unclear whether some of the leaked data might have been repackaged from earlier incidents, researchers insist that this leak is largely new. Lawrence Pingree, vice president at cybersecurity firm Dispersive, explained that such datasets are often circulated and resold on the dark web—sometimes bundled with other leaks, sometimes offered piecemeal. 'Whether it's a repackaged leak or not, 16 billion records is a huge number,' Pingree said. 'This kind of data is valuable precisely because it is so often misused.' The breach underscores how widespread the threat of credential theft has become, with attackers targeting social media platforms, corporate portals, developer tools, and VPN services alike. In response, experts urge users to adopt better security hygiene. Basic protections include running antivirus scans to detect infostealers, checking dark web exposure via tools like Google One's 'Dark Web Report,' and crucially, using strong and unique passwords for every service.

Mint

an hour ago

• Mint

Alphabet's Google tries to appease EU with changes to search result rankings: Report

Alphabet's Google has reportedly put forward additional concessions to address concerns raised by European Union regulators, aiming to avoid a significant antitrust fine under the bloc's new digital competition rules. According to documents reviewed byReuters, the U.S. tech giant is attempting to appease the European Commission by adjusting how rival services are displayed in its search results. The move comes in response to formal charges filed three months ago, accusing Google of favouring its own platforms, such as Google Shopping, Hotels, and Flights at the expense of competitors, in violation of the Digital Markets Act (DMA). You may be interested in The DMA, which came into force earlier this year, outlines strict obligations for so-called 'gatekeeper' platforms to curb anti-competitive behaviour and offer consumers broader choices. Under Google's revised proposal, a selected vertical search service (VSS), chosen based on objective and non-discriminatory criteria would be prominently featured in its own dedicated box at the top of the results page. This box would mirror the design and features of Google's own modules and contain three direct links to offerings in categories like hotels, restaurants, transport, and airlines. Other VSS providers would still be listed further down in the search results, but would not benefit from a similarly prominent display unless users click to access them. Despite the proposal, the company has maintained that it disagrees with the Commission's preliminary conclusions. 'We do not agree with the (Commission's) preliminary findings' position but, on a without prejudice basis, we want to find a workable solution to resolve the present proceedings,' the documents noted. The European Commission has scheduled a meeting on 8 July to gather feedback from competing firms. Several of Google's rivals, who declined to be named ahead of the discussion, expressed scepticism over the effectiveness of the proposed changes. They argue that the measures still fall short of delivering a genuinely level playing field. The outcome of these negotiations could set a crucial precedent for how Big Tech operates under the EU's ambitious digital regulatory framework. (With inputs from Reuters)