Perplexity in talks with phone makers to pre-install Comet AI mobile browser on devices
The move could significantly boost Perplexity's reach by capitalising on browser "stickiness", where users tend to stick with browser apps that are pre-installed or set as default on their devices, potentially driving habitual use of the company's AI tools.
"It's not easy to convince mobile OEMs to change the default browser to Comet from Chrome," Srinivas said, referring to original equipment manufacturers and highlighting the challenge of user inertia on mobile platforms.
Comet, currently in beta and available only on desktops, integrates Perplexity's AI directly into web browsing, allowing users to ask questions about personal data like emails, calendars, or browsing history, and even perform tasks such as scheduling meetings or summarising webpages.
Perplexity aims to target "tens to hundreds of millions" of users next year after stabilising the desktop version for a few hundred thousand initial testers, Srinivas said.
Its efforts reflect a broader industry shift toward browsers with agentic AI capabilities, ones that need minimal human intervention to make decisions and achieve specific tasks.
Reuters reported earlier this month that OpenAI is developing its own agentic AI browser, which could automate complex tasks such as booking travel or managing finances.
As of last month, Google's Chrome had a market share of about 70% in mobile devices, while Apple's Safari and Samsung's browsers together commanding another 24%, according to Statcounter data.
Bloomberg News reported in June that Perplexity was in talks with Samsung Electronics and Apple to integrate its AI search capabilities into their devices, potentially enhancing assistants like Bixby or Siri.
Perplexity has completed a $500 million investment round, which valued it at $14 billion earlier this year. Its investors include Accel, Nvidia, Jeff Bezos and former Google CEO Eric Schmidt.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Time of India
29 minutes ago
- Time of India
Tower growth may hit a plateau as 5G nears completion
Live Events (You can now subscribe to our (You can now subscribe to our Economic Times WhatsApp channel Mumbai: As telecom companies such as Reliance Jio and Bharti Airtel near completion of 5G network expansion and curtail capital expenditure and tower rentals fall, tower companies are likely to see flattish growth in the next five years, with Vodafone Idea 's 5G rollouts and 4G coverage expansion likely to offer short-term support, according to new tower additions fell to 25,000 in 2024-25, the lowest since 2017-18, and are expected to dip further in this financial year, according to data from ICICI Securities. It comes after industry towers saw a deceleration in growth to 3.5% in the previous fiscal after a robust 8.1% annual expansion since additions of 5G BTS, or radios, which constitute a large part of network capex, also nosedived, to 8,000, just into the third year of 5G rollouts, from 141,000 in 2022-23. This indicates a conspicuous drop in 5G capex despite 5G BTS accounting for only 15.8% of the total. In contrast, 4G BTS' plunged after seven-eight years of analysts expect a lull for telecom tower companies in the next five years, as significant tower or BTS additions are likely only when telecom operators buy new spectrum bands to deploy 6G."5G is being deployed only on two spectrum bands - 700MHz and 3300MHz - implying industry BTS/tower should start stabilising," ICICI Securities said in a research report. "In the future, as 4G gets re-farmed, existing 4G spectrums and BTS would be deployed for 5G; therefore, keeping BTS/tower unchanged."Out of the nearly 800,000 telecom towers in India - the second largest in the world after China - about 275,000 are run by Brookfield-backed Altius Telecom Infrastructure Trust (formerly Data Infrastructure Trust), which counts telecom market leader Reliance Jio as its largest tenant. Around 250,000 are run by Indus Towers , which is majority owned by the country's second-largest telecom operator, Bharti Airtel, while other major tower companies include GTL Infrastructure , with 26,000 towers, and Ascend Telecom, with about 18,600 telecom sector capex has materially declined since 2022-23, when 5G rollouts started. While Reliance Jio was most aggressive in the first two years of rollout, Bharti Airtel's capex was calibrated over four years. Meanwhile, Vodafone Idea, which has just started its 5G rollout and 5G network expansion, is unlikely to move the needle on tower additions to that extent."After a good rollout by Bharti (Airtel) in last couple of years, we expect Bharti's tower adds to be in the range of 5-8K (down from an estimated 15,000-20,000 in the past two years)," BofA Securities said in a recent report. "We also expect industry tower adds to slow as 5G would require lesser traditional towers/more small cells, even as business from Vodafone Idea will remain strong for next 2-4 quarters."Credit research firm India Ratings has given a weak outlook for tower companies in this fiscal, as tenancies have continued to decline and margins have remained flat. Tower rentals fell sharply to '69,000 per tower per month in 2024-25 from '85,000 in 2021-22, data showed.
Economic Times
an hour ago
- Economic Times
Be the energy sAIviour: India's green power can fuel the world's AI future
As AI continues its march, the demand for energy-intensive AI processing is skyrocketing. India, with its abundant RE resources and burgeoning AI ecosystem, is well-positioned to become a global leader in green energy-powered AI processing. AI computing needs large-scale data centres, or hyperscale data centres. These require reliable green power at a scale and speed. Global data centre energy requirements are projected to reach 4,000 TWh by 2030, accounting for 5% of global electricity demand. Estimates suggest that training a single AI model can consume up to 284,000 kWh. Renewable power plants have the shortest commissioning timelines due to their modular nature. India has set ambitious RE targets, aiming to generate 50% of its electricity from non-fossil fuels by 2030. With over 300 days of sunshine per year and generous wind speeds, India has immense solar energy potential, which can be harnessed to power AI processing. India's true strength lies in local green energy champions, who have the capacity to implement world-class projects. The renewables industry is backed by a modern single national grid and a dynamic regulatory framework. India's AI ecosystem is growing rapidly, with over 1,000 AI startups. Moreover, 20% of global AI talent resides in India, making it an attractive location for AI companies. The AI market size in India is expected to reach $7.8 bn by 2025, driven by increasing demand for digital services, ecommerce and cloud computing. India's data centre market is growing rapidly, driven by increasing demand for digital services, ecommerce and cloud computing. According to a report by MarketsandMarkets, the Indian data centre market is expected to reach 1,432 MW of installed capacity by 2025, growing at a CAGR of 21.1%. By 2030, the market is projected to reach 3,243 MW of installed capacity, with a growth rate of 15.6%.The needs of a data centre are distinct from other infrastructure projects. They are designed and built for the highest standards of reliability, modularity, redundancy and security. The ability of the data centre to provide backup power with redundancy is also critical for its though diesel generators are the dominant backup technology for data centres, green technologies such as batteries (6-hour backup) and hydrogen fuel cells (48-hour backup) are increasingly being explored to reduce carbon intensity. With water being a significant barrier to the expansion of hyperscale data centres, intelligent integration with technologies such as fuel cells that produce water as a by-product could be an interesting proposition. Google's AI-powered data centre in Hyderabad is powered by 100% RE, setting a precedent for sustainable AI processing. Microsoft's AI-powered data centre in Pune uses solar power for primary energy needs, further demonstrating the potential for green energy-powered AI processing. The Make in India initiative encourages the development of green data centres and AI infrastructure, providing a supportive policy framework for the growth of this sector. To ensure success in the AI data centre sector, India should promote a net- zero hyperscale data centre policy, aiming to position itself as the world's most sustainable data centre destination. The country should aim to unlock a significant number of carefully selected sites that can provide 24x7 green power with appropriate green backup power India should nurture skilled talent capable of building cutting-edge data centres with high levels of autonomy and security. Innovations for efficient configuration of data centres that minimise energy and water requirements should be promoted through publicly funded pilots. Finally, India should build global trust in the integrity of data protection on Indian soil with appropriate policy, regulatory and outreach initiatives. India's clean energy strengths and growing AI ecosystem make it an ideal location to become a world leader in green energy-powered AI processing. By leveraging its RE strengths and addressing the challenges of AI data centres proactively, India can reduce its carbon footprint, enhance its reputation as a hub for sustainable business and innovation, and drive economic growth and job creation in the clean energy and AI sectors.
India.com
2 hours ago
- India.com
Google Unveils Key Initiatives To Boost India's AI Capabilities At I/O Connect 2025
Bengaluru: Google unveiled a range of initiatives aimed at propelling India's AI capabilities and empowering developers at Google I/O Connect India 2025 in Bengaluru on Wednesday. During the event, the key announcement was the localisation of onshore processing for Gemini 2.5 Flash, enabling faster, low-latency AI performance for Indian developers—especially in regulated sectors such as healthcare, finance, and governance. 'Indian developers are literally writing the next chapter of India's success story, using AI capabilities to build real-world applications that are reaching millions of businesses and people across India and the world,' said Manish Gupta, Senior Director for India and APAC at Google DeepMind. The company also emphasised its continued commitment to India's developer ecosystem, revealing that the Play and Android platforms contributed ₹4 lakh crore in revenue and generated 35 lakh jobs in 2024. Google further introduced agentic AI tools and optimised templates in Firebase Studio, allowing developers to build and deploy full-stack AI applications more quickly and efficiently. In a major move for mobile commerce, Google announced expanded access to over 250 million mapped places globally and launched India-specific pricing for Google Maps UI components. Additionally, AI-powered summaries in the Places API will help developers create location-based services tailored specifically for Indian users. To further India's AI ambitions, Google is collaborating with three India AI Mission startups—Sarvam, Soket AI, and Gnani—on Make-in-India AI models based on Gemma. Sarvam's recent model, 'Sarvam-Translate,' exemplifies this homegrown innovation. In partnership with IIT Bombay's BharatGen, Google is also advancing ASR (Automatic Speech Recognition) and TTS (Text-to-Speech) models for Indic languages. To support India's thriving game development community, Google launched the 'Google Play x Unity Game Developer Training' program, in collaboration with Unity and GDAI. Targeted at 500 Indian developers initially, the program offers over 30 hours of specialised training across various game development roles. Highlighting grassroots innovation, the Gen AI Exchange Hackathon encouraged developers to apply their AI skills to real-world challenges. During the event, startups such as Sarvam, InVideo, Glance, and Nykaa showcased impactful solutions built using Google's AI models.
