Home Blog Page 35

Oracle to flex muscles with rivals with Google’s AI partnership

Naushad K. Cherrayil
-
0
Oracle to flex muscles with rivals with Google’s AI partnership
  • Users can tap into Gemini through Oracle’s Universal Credits system, meaning there’s no jarring shift in procurement or billing
  • As top giants jostle for dominance, enterprises stand to benefit from an AI ecosystem that’s more open, more capable, and quicker to deliver tangible productivity gains.

Oracle is clearly doubling down on its generative AI ambitions, and the company’s latest move—joining forces with Google Cloud to integrate Gemini models into Oracle Cloud Infrastructure—turns heads for good reason.

By making Google’s advanced Gemini AI, starting with Gemini 2.5, available on the OCI Generative AI service, Oracle doesn’t just add a new tool; it turbocharges its appeal to enterprises itching for multi-cloud flexibility.

What’s in it for Oracle customers? For one, easy access: users can tap into Gemini through Oracle’s Universal Credits system, meaning there’s no jarring shift in procurement or billing. But this integration is about more than simplicity.

The shared roadmap brings far more than text generation—think images, video, and audio, plus sector-focused models like MedLM for healthcare.

And when Gemini gets deeply woven into Oracle’s business-critical Fusion Cloud Applications—think finance, HR, and supply chain—the productivity upside for end users could be huge.

Oracle’s AI bet

The numbers back up Oracle’s AI bet. In the fourth quarter of fiscal 2025, Oracle’s total cloud revenue soared 27 per cent year over year to $6.7 billion. That’s not a fluke—it’s a signal that enterprises are buying into Oracle’s AI-first pitch.

By supporting leading AI models from not just Google, but also Cohere, Meta (with Llama), and xAI’s Grok, Oracle gives its clients options and insulates them from vendor lock-in. This multi-model buffet does more than just drive OCI traffic; it catalyses AI adoption across a broad range of use cases in Oracle’s applications, paving the way for durable growth.

For Google, meanwhile, this partnership is an entry into Oracle’s vast enterprise client base, extending Gemini’s reach and deepening the kind of cross-cloud integration that started with Database@Google Cloud.

There’s a clear ambition here: help businesses leap from just using AI to harnessing the “agentic” AI—systems that proactively automate and orchestrate tasks—expected to define the next wave of enterprise productivity.

Cross-cloud partnerships

The momentum should carry Oracle’s growth forward. Analysts anticipate mid-to-high teen revenue growth rates for fiscal 2026 and 2027—figures that seem plausible if the AI flywheel keeps spinning.

Zooming out, Microsoft and Amazon, two titans in the cloud AI game, are not standing still. Microsoft’s AI-laced Azure platform, thanks to deep partnerships with OpenAI, Nvidia, and Anthropic, notched $75 billion in revenue last fiscal year, growing 34 per cent.

With an unmatched global cloud network, robust Copilot uptake, and relentless cash generation, Microsoft’s scale keeps it at the tip of the AI spear.

Amazon, meanwhile, leverages AWS’s 32 per cent share of the global cloud market—fueled by a jaw-dropping $100 billion in data center investments and homegrown AI chips like Trainium 2, which offer price-performance rivaling Nvidia.

Amazon Bedrock’s model buffet, featuring Claude and Llama, plus double- and triple-digit AI revenue growth rates, signal that Amazon’s AI momentum isn’t likely to fade.

Where does this leave Oracle? In a three-way race where differentiators like flexibility, cross-cloud partnerships, and depth of embedded AI for business workflows could make or break market share.

As these giants jostle for dominance, enterprises stand to benefit from an AI ecosystem that’s more open, more capable, and quicker to deliver tangible productivity gains.

Cohere secures $500m in oversubscribed funding round

TCN Bureau
-
0
Cohere secures $500m in oversubscribed funding round
  • Funding round led by Radical Ventures and Inovia Capital, and saw enthusiastic participation from existing investors like AMD Ventures, NVIDIA, PSP Investments, Salesforce Ventures and Healthcare of Ontario Pension Plan.

Cohere, the Canadian artificial intelligence startup, has successfully closed an oversubscribed funding round, raising a remarkable $500 million and pushing its valuation to $6.8 billion.

The round was led by Radical Ventures and Inovia Capital, and saw enthusiastic participation from existing investors like AMD Ventures, NVIDIA, PSP Investments, and Salesforce Ventures.

Notably, new backers such as the Healthcare of Ontario Pension Plan (HOOPP) also joined the roster, signaling strong confidence in Cohere’s trajectory.

With these fresh funds, Cohere is set to accelerate its efforts in agentic AI—paving the way for smarter, more autonomous AI systems that can transform enterprise operations.

A competitive edge

The strategic direction comes on the heels of several high-profile collaborations with industry leaders including Oracle, Dell, Bell, Fujitsu, and SAP.

Founded in 2019, Cohere has quickly carved out its place in the competitive AI landscape, offering language models and AI tools that automate complex tasks across a variety of sectors.

The company stands shoulder to shoulder with US AI giants like OpenAI and Google, constantly striving to deliver secure, high-performing language technologies that give clients a competitive edge.

Adding more fuel to its momentum, Cohere has bolstered its leadership team with two major hires. Joelle Pineau, previously vice-president for AI research at Meta and a renowned professor at McGill University, steps in as Cohere’s first chief AI officer.

Pineau brings extensive experience from her tenure as lead of Meta’s Fundamental AI Research (FAIR) team, positioning Cohere for more cutting-edge innovation.

Joining her is Francois Chadwick, Cohere’s newly appointed chief financial officer. With a background that spans executive roles at Uber and a partnership at KPMG US, Chadwick will spearhead finance and business operations as the company gears up for rapid expansion and global reach.

Solidifying its position

Aidan Gomez, co-founder and CEO of Cohere, shared his enthusiasm: “Cohere is becoming the world’s chosen partner for integrating AI into their critical industries. We are at a pivotal moment in accelerating the delivery of secure AI that empowers enterprises worldwide, and we’re excited to enter this new phase of expansion alongside our partners.

“We’re thrilled to welcome Joelle and Francois, whose experience, talent and insight will be instrumental in continuing Cohere’s growth.”

Echoing this optimism, Jordan Jacobs, co-founder and managing partner at Radical Ventures, emphasised Cohere’s unique approach: “Cohere is fulfilling that promise by building privacy-first, cloud-agnostic models and agentic AI applications that are driving extraordinary productivity gains and ROI to blue-chip enterprises, businesses and governments worldwide. We are co-leading this investment round because this is just the beginning for Cohere.”

With fresh capital, a reinforced leadership team, and an ambitious outlook, Cohere is poised to solidify its position as a key player in the evolving world of enterprise AI.

Who’s winning the AI battle for user attention in 2025?

Naushad K. Cherrayil
-
0
Who’s winning the AI battle for user attention in 2025?
  • ChatGPT is still the sun at the centre: it had 46.6b visits—more than twice all the next nine chatbots combined.
  • ChatGPT may still dominate by sheer size, but rivals like Grok, Gemini, Perplexity, and Claude are nipping at its heels by refining user experience and solidifying loyalty.

Each month, over 60,000 people search for the “best AI chatbot”—and that number’s only going up.

We’re at the epicentre of an AI chatbot explosion, and the sheer pace is dizzying. It’s like suddenly finding yourself at the birth of the cosmos: one spark of innovation gives way to countless new stars.

What started with a handful of promising tools has now swelled into a bustling universe with thousands of chatbots, each vying for their turn in the spotlight.

It’s easy to get lost in all these choices. Whether you’re a marketer, teacher, entrepreneur, or just chatbot-curious, the quest for what actually works can feel overwhelming.

You’re not alone if you’ve hopped onto a listicle, only to find yourself more confused than when you started. People want substance, not just more options. Many are looking for viable ChatGPT alternatives. But is ChatGPT still the king, or are broader ambitions taking root elsewhere?

Reflecting on these questions, Sujan Sarkar, Co-Founder at onelittleweb.com, described the current moment as “an AI chatbot explosion that’s reshaping how the world interacts with technology.” Sarkar’s team analysed more than 10,500 AI tools between August 2024 and July 2025—a Herculean effort to map this rapidly expanding world.

Leaders and laggards in the AI race

Over the last year, the top 10 AI chatbots drew an incredible 55.88 billion visits. That’s a 123 per cent increase over the previous year! ChatGPT is still the sun at the centre: it had 46.6 billion visits—more than twice all the next nine chatbots combined.

But look out, because not all stars rise at the same speed. Grok, once obscure, had just 51,000 visitors last year but rocketed to a mind-blowing 687 million in 12 months—an increase of more than 1,340,000 per cent! This kind of growth is almost unheard of in any tech sector.

DeepSeek’s trajectory is more of a rollercoaster. It climbed quickly to 520 million visits in February 2025 but then dropped 39 per cent by July, showing how fickle user attention can be.

Meanwhile, Google’s Gemini became a heavy-hitter, growing 156 per cent year-over-year to reach 1.7 billion visits, buoyed by better integration with Google’s ecosystem.

Claude, another rising star, claims the most engaged users: people spend an average of almost 17 minutes per session, and it’s also racked up more than 40 million app store reviews—a testament to its strength in the mobile market.

Perplexity and Claude, while still behind ChatGPT in raw numbers, are quietly building powerful, loyal user bases. Both have more than doubled traffic, each reaching the billion-visit milestone.

On the flip side, some players are losing steam: Poe saw a drop of 46 per cent in visits, and both Mistral and Meta AI are stuck in the slow lane despite relaunches and fresh branding.

All told, these ten tools account for about 59 per cent of all AI-driven web traffic. In other words, the biggest planets still rule the solar system, but the smaller ones are catching up fast.

What lit the fuse?

The dramatic jump in total traffic—from 25 billion to almost 56 billion visits year over year—didn’t happen by accident.

The hottest period for growth was between March and July 2025, with May alone smashing records at 6.4 billion visits (compared to 2.4 billion last spring).

March 2025 was an inflection point. It saw an eye-popping 141 per cent surge in visits, kicked off by platform upgrades to ChatGPT, DeepSeek’s media blitz, and the arrival of new chatbots with multimodal chops (think: handling text, images, even audio).

“Success always depends on what you measure,” said Sarkar. DeepSeek, for instance, ranks fifth in their overall model, but rockets to second if you just count annual visits. That’s a reminder: user engagement, session duration, and retention paint very different pictures than raw traffic alone.

Just as interesting—who keeps users hooked the longest? Claude reigns here, with nearly 17 minutes per session. Grok and ChatGPT also keep people engaged for more than 15 minutes on average.

There’s no single “best” AI chatbot—at least, not for everyone. ChatGPT may still dominate by sheer size, but rivals like Grok, Gemini, Perplexity, and Claude are nipping at its heels by refining user experience and solidifying loyalty.

If anything, the metrics show that how you define “winning” changes depending on where you look: is it about traffic? Engagement? Rapid growth?

Whatever your perspective, there’s no denying that this is an era of AI abundance. And one thing’s for sure: the chatbot universe is still expanding—and the race for attention has only just begun.

OpenAI cautions users: GPT-5 is powerful but not infallible

Mohammed Fahad
-
0
OpenAI cautions users: GPT-5 is powerful but not infallible
  • Despite ongoing improvements, ChatGPT should remain a “second opinion” tool—not a definitive source of truth.

OpenAI’s latest language model, GPT-5, has taken a significant leap in power and precision compared to earlier versions, but users are being cautioned not to place blind trust in its responses.

Nick Turley, Head of ChatGPT, recently underscored the fact that despite ongoing improvements, ChatGPT should remain a “second opinion” tool—not a definitive source of truth.

Turley, speaking with The Verge, was candid about the persistent issue of AI hallucinations. Even with advances in the underlying technology, GPT-5 occasionally generates information that appears convincing yet is factually incorrect.

OpenAI’s own assessments indicate that the model still produces wrong answers roughly 10 per cent of the time—an improvement, but not perfection.

Turley highlighted the complexity of the task: “Achieving total reliability is a massive challenge,” he said.

He made it clear that as long as language models lag behind human experts in their accuracy across all domains, OpenAI will continue to suggest users double-check the AI’s advice. “Until we are provably more reliable than a human expert across all domains, we’ll continue to advise users to double-check the answers,” Turley noted.

For now, ChatGPT is best seen as a supplement—an extra set of eyes on complicated questions, not the only authority.

Why errors still happen

Large language models like GPT-5 generate answers by recognising patterns in enormous text datasets. This allows them to excel at natural, humanlike conversation, but it also means they can present incorrect facts on topics that aren’t well-represented in their training data, or even invent plausible-sounding details that aren’t true.

To help users catch any slip-ups, OpenAI has equipped ChatGPT with search functionality—making it easier to verify answers by cross-referencing with trustworthy external sources.

Turley was optimistic about eventual solutions but tempered expectations by admitting that eliminating hallucinations will take time: “I’m confident we’ll eventually solve hallucinations, and I’m confident we’re not going to do it in the next quarter.”

Despite these challenges, OpenAI isn’t slowing its ambitions. Reports indicate the company is working on launching its own web browser, while CEO Sam Altman has even made tongue-in-cheek remarks about potentially buying Google Chrome if it ever came onto the market. Clearly, OpenAI intends to expand well beyond chatbots and continue shaping the way people interact with information online.

Foxconn ramps up iPhone 17 production at India plant

Ana Mary Joseph
-
0
Foxconn ramps up iPhone 17 production at India plant

  • Foxconn fills critical technical gaps by bringing in experts from Taiwan and elsewhere after the unexpected departure of several Chinese engineers.
  • For the year ended March 31, 2025, Apple assembled 60% more iPhones in India, representing a market value of about $22b.

Foxconn, the Taiwanese electronics behemoth and Apple’s principal manufacturing partner, has officially kicked off production of the iPhone 17 at its new, state-of-the-art facility in Bengaluru.

After months of anticipation and a massive investment of around $2.8 billion (approximately Rs25,000 crore), this development signals a new chapter for India’s role in Apple’s global supply chain.

The iPhone 17 series launch dates have been rumoured to be in early September, likely on September 9.

The Devanahalli-based unit now stands as Foxconn’s second-largest iPhone plant outside China, supplementing output from the company’s long-established Chennai factory. Both locations are currently humming with activity as they manufacture the latest iPhone 17 series.

The achievement echoes Foxconn’s recent momentum after last year’s local iPhone 16 production, which had also preceded the phone’s global and Indian debuts.

Sources familiar with internal operations suggest the Bengaluru plant has quickly rebounded from earlier hiccups, including the unexpected departure of several Chinese engineers. Foxconn’s ability to fill critical technical gaps by bringing in experts from Taiwan and elsewhere has been key to keeping production on track.

Apple’s bet on India

Apple is now doubling down on India as a manufacturing powerhouse. Projections estimate iPhone output will reach a staggering 60 million units this calendar year—significantly up from the 35–40 million units produced during 2024–25.

For the year ended March 31, 2025, Apple assembled 60 per cent more iPhones in India, representing a market value of about $22 billion.

Apple’s CEO Tim Cook has publicly recognised India’s rising profile. Following the company’s July 31 financial results, Cook highlighted that most iPhones sold in the US in June 2025 were made in India.

During the June quarter’s earnings call, he noted that all iPhones shipped to the US from India during that timeframe—a major shift in Apple’s supply chain dynamics.

Scaling production to meet demand

S&P Global’s data reveals that iPhone sales in the US reached 75.9 million units in 2024. With Indian exports of iPhones hitting 3.1 million units in March 2025 alone, Apple will need to dramatically ramp up capacity at its Indian facilities or reroute more devices intended for Indian consumers to international markets.

Meanwhile, Apple’s footprint in India’s burgeoning smartphone market continues to expand. During the first half of 2025, sales climbed 21.5 per cent year-on-year to 5.9 million units, led by the iPhone 16 as the best-selling model.

The June quarter further cemented Apple’s momentum, with YoY shipments soaring nearly 20 per cent, giving the brand a 7.5 per cent share of India’s smartphone market.

Despite these gains, Chinese OEMs continue to dominate the broader Indian market. IDC’s data places Vivo at the top with a 19 per cent share for the June quarter, underscoring the competitive landscape Apple faces.

DeepSeek and the global shift in open-source AI revolution

Naushad K. Cherrayil
-
0
DeepSeek and the global shift in open-source AI revolution

  • DeepSeek proved that transformative innovation need not be limited to one geography, language, or company—and that knowledge, once shared freely, multiplies in unpredictable ways.
  • The promise of open AI cracked open a spirited, sometimes fierce debate: which is safer and wiser—the freedom of the crowd, or the guardrails of the few?
  • Open-source AI isn’t just a technological leap; it is compressing costs, catalysing new forms of innovation, and shifting where value—economic, societal, and political—will flow in the coming decades.
  • Data, integration, governance—these will be the battlegrounds of tomorrow’s AI economy.

Once upon a time, in the looming shadow of Silicon Valley, artificial intelligence looked like a lush private garden—vibrant with promise, but walled off for the privileged few.

For years, American powerhouses like OpenAI and Google poured billions into crafting towering models, all secretive code and legal barricades. Many onlookers assumed only the largest, wealthiest tech giants could truly orchestrate the future of AI.

Across the Pacific, however, a spark of rebellion glimmered. DeepSeek, a young and audacious startup from China, dared to imagine a world where AI would be a shared resource rather than a fortified treasure.

Its mission was nothing short of radical: break open the locked gates of AI and let the light spill out for everyone.

Fueled by this vision, DeepSeek’s engineers persisted with relentless energy, building models like DeepSeek-R1 and DeepSeek-V3—rivalling their Western counterparts not just in raw intelligence, but in core efficiency. They obsessed over optimisation, turning every dollar and hour into sharper algorithms and smaller, faster footprints.

But DeepSeek didn’t just keep its research inhouse or erect any artificial boundaries. In a move that startled the world, it published its models—without restrictions, usage caveats, or hidden fees.

No “research only” fine print. Coders in Mumbai, creatives in Nairobi, and entrepreneurs from São Paulo to Sydney suddenly found themselves empowered with best-in-class AI tools, liberated from the corporate stronghold of American tech behemoths.

DeepSeek’s open gamble

The impact was immediate. DeepSeek-R1, to the world’s astonishment, shot above even ChatGPT in the US app charts for a time. The old status quo began to crack. Tech giants scrambled: some released hurried, less-restricted models, trying to ride DeepSeek’s momentum.

Markets shivered as investors began to worry that open-source disruption would erode the fat monopolies of the past.

By stripping away all the old, costly hurdles—expensive licenses, tight legal leashes—DeepSeek re-leveled the playing field. Now, with models trained for just $5 million—a fraction of the hundreds of millions spent by rivals—startups, hobbyists, and researchers everywhere could join the AI race in earnest.

That leap in efficiency threatened to trigger the so-called “Jevons paradox” in AI: as costs plummet and access widens, adoption will surge in previously unimaginable ways.

DeepSeek’s open gamble forced responses even from the incumbents. OpenAI, after a long hiatus, finally released an open model—evidence that the open-source wave was becoming inescapable. China’s rise via DeepSeek was not just symbolic; it marked a real-world shift in the balance of technological power and innovation.

But this new era of openness also brought dilemmas. Policymakers and security experts sounded alarms—could free and universal access to potent models enable cyberattacks, disinformation spree, or even mass manipulation?

Catalysing new forms of innovation

The promise of open AI cracked open a spirited, sometimes fierce debate: which is safer and wiser—the freedom of the crowd, or the guardrails of the few?

The genie, of course, refused to return to the bottle. DeepSeek proved that transformative innovation need not be limited to one geography, language, or company—and that knowledge, once shared freely, multiplies in unpredictable ways. AI, once a playground for the privileged, had become everyone’s tool.

As new voices joined the fray and new ideas blossomed, the future of intelligence suddenly belonged not to an elite, but to the world at large.

Open-source AI isn’t just a technological leap; it is compressing costs, catalysing new forms of innovation, and shifting where value—economic, societal, and political—will flow in the coming decades.

Data, integration, governance—these will be the battlegrounds of tomorrow’s AI economy. For now, with DeepSeek’s revolution as a beacon, humanity stands on the threshold of an age where brilliance is a birthright, not a privilege.

1...343536...267Page 35 of 267