Рубрика: AI

SoftBank founder Masayoshi Son said he was emotional about parting with the company’s entire NVIDIA stake, but insisted the sale was unavoidable as the group accelerates its push into AI.

Speaking at the FII Priority Asia forum in Tokyo on Monday, Son addressed SoftBank’s November disclosure that it had sold its NVIDIA shares for $5.83 billion. The tech billionaire said the decision wasn’t driven by doubts about the chipmaker but by SoftBank’s need to finance major new AI projects.

“I don’t want to sell a single share. I just had more need for money to invest in OpenAI and other projects,” Son said. “I was crying to sell NVIDIA shares.”

SoftBank has spent the year ramping up its AI ambitions, pouring resources into initiatives including the massive Stargate Project data centres and the acquisition of US chip designer Ampere Computing.

The company is also preparing to expand its backing of OpenAI, with a potential increase depending on the startup’s performance and valuation in future funding rounds, a person familiar with the discussions previously told CNBC.

Son has repeatedly positioned OpenAI at the centre of SoftBank’s next phase, declaring earlier this year that the group is “all in” on the ChatGPT maker and predicting it could eventually become the world’s most valuable company.

That conviction has already delivered financial returns. SoftBank reported that second-quarter net profit more than doubled to 2.5 trillion yen ($16.6 billion), supported by gains tied to its OpenAI stake.

The company’s aggressive posture comes as investors debate whether the AI sector is overheating. Son dismissed those concerns on Monday, saying people warning of an AI bubble are “not smart enough.”

He argued that advancements in “super [artificial] intelligence” and robotics will eventually create at least 10% of global GDP, easily justifying the trillions of dollars currently flowing into the technology.

Despite selling NVIDIA, now one of the world’s most valuable semiconductor firms, Son made clear that the move was purely strategic. SoftBank, he suggested, is reallocating capital not away from AI, but deeper into it.

The post SoftBank’s Son Says He ‘Was Crying’ Over NVIDIA Stake Sale, But Needed Cash to Fund OpenAI appeared first on Analytics India Magazine.

Hyderabad-based Cyient Semiconductors has been qualified for a pivotal mandate in the ₹4,500 crore modernisation of the Semiconductor Laboratory (SCL) in Mohali, marking a significant advancement in India’s efforts to expand domestic semiconductor manufacturing capacity under the India Semiconductor Mission (ISM).

The qualification positions Cyient Semiconductors to supply and validate key technology IPs—RF-CMOS, BCD (HV LDMOS), and CMOS Image Sensor (CIS)—which will form the backbone of SCL’s enhanced 8-inch fabrication line. These mature-node technologies are crucial to several fast-growing sectors, including industrial systems, automotive electronics, energy management, imaging, sensing and connectivity.

The government’s larger objective for SCL’s overhaul is to create accessible fabrication capability for startups, academia, and strategic sectors, while reducing India’s heavy dependence on imported semiconductors. Cyient Semiconductors’ role directly contributes to these goals by modernising the process platforms that will power the upgraded fab.

Calling the development a “proud moment,” Krishna Bodanapu, executive vice-chairman and managing director of Cyient Limited, said the qualification reflects the company’s engineering depth and turnkey execution strengths. “Our collaboration with SCL will accelerate India’s semiconductor self-reliance by delivering highly relevant, high-value silicon solutions in digital, analogue mixed-signal and power domains,” he said.

Suman Narayan, CEO of Cyient Semiconductors, noted that the selected technologies align with the company’s strongest domains. He described the project as both a responsibility and an opportunity to contribute meaningfully to a national mission.

The updated technologies could eventually enable SCL to support applications ranging from smart energy systems and industrial control to imaging, sensing, and low-power IoT devices, areas where mature-node processes remain highly relevant.

Earlier this year, Cyient Semiconductors and MIPS, a global leader in RISC-V processor technology, teamed up to develop specialised chips for power management, industrial robotics, and automotive applications.

Through the partnership, they aim to address real-time, safety-critical applications, power delivery, and compute efficiency in automotive, industrial, and data centre markets, with a focus on motor control and data centre power delivery platforms, according to a statement.

The post Cyient Semiconductors Qualifies to Supply Core Tech for ₹4,500 Cr SCL Mohali Modernisation appeared first on Analytics India Magazine.

Forrester, the global research and advisory firm, closed its 2026 Predictions India event in Bengaluru on December 5.

The firm outlined a set of forecasts that signal stricter oversight of AI investments, stronger governance expectations, and a shift back to core technology fundamentals.

Among the eight predictions the firm made, it expects global workflow automation giant ServiceNow to acquire Boomi, the integration platform provider.

“We’ve seen the iPaaS [Integration Platform as a Service] market being a prime acquisition target for the last seven years,” said Leslie Joseph, principal analyst at Forrester, in a keynote. He pointed towards how Salesforce made a similar acquisition of MuleSoft in 2018.
“ServiceNow and Boomi have made interesting moves over the last two years — to bring their products, portfolios, and GTMs [go-to-market strategies] closer,” said Joseph, highlighting it wouldn’t be a surprise if the acquisition takes place.

Joseph said the prediction reflects a broader shift away from an app-centric enterprise architecture toward one in which data, domain logic, and AI capabilities sit outside individual applications and are coordinated through agents and orchestration layers.

In this model, iPaaS platforms become the control point that connects and governs how work flows across systems and AI agents.

Because ServiceNow is trying to strengthen its position in these orchestration and governance layers, he said, acquiring Boomi fits the direction of the market, with iPaaS moving to the centre of enterprise architecture.

Currently, ServiceNow and Boomi have a publicly recognised strategic partnership, where the latter offers API management and integration solutions specifically targeting ServiceNow’s platform.

The predicted acquisition underscores Forrester’s advice for Indian enterprises to rationalise their integration stacks. The firm stated, “As integration platforms become the control plane for AI orchestration, Indian enterprises will need to rationalise overlapping iPaaS and workflow tools.”

Other 2026 Predictions: AI Governance, Technical Debt, Cloud Sovereignty

Another major prediction was that a chief information officer of a Global 1000 company would declare “technical debt bankruptcy,” highlighting that legacy estates consume nearly 60-80% of IT budgets.

For Indian enterprises with ageing ERP systems, Forrester expects more outsourcing of legacy operations and increased investment in cloud-native platforms.

Another prediction stated that “20% of Indian brands will miss cost targets or lose trust in vendors after overpromising on AI.” This was attributed to the customer service leaders failing to realise the expected cost savings from AI.

Leaders will face pressure to ‘renegotiate contracts and reimagine stagnant service categories to restore trust and efficiency.’

On a similar note of the misrepresentation of generative AI’s capabilities, Forrester stated in another prediction that it will lead to one Fortune 500 company suing a B2B provider. “As Indian enterprises scale AI-driven marketing and sales, governance lapses will result in misinformation and legal disputes.”
Other predictions include Indian CIOs being drawn in to fix AI deployments launched without adequate governance. Forrester expects a quarter of CIOs to be tasked with bailing out business-led AI projects that were running ahead of technical and risk checks, particularly in sectors such as banking and insurance, where adoption has accelerated.

The firm also said AI and digital sovereignty will drive a double-digit year-over-year growth for the private cloud ecosystem.

“India’s data localisation laws and the National Quantum Mission will amplify demand for sovereign hybrid cloud architectures,” Forrester stated, adding that enterprises will prioritise private cloud for sensitive workloads and AI model training to mitigate geopolitical risks and ensure compliance with emerging sovereignty mandates.

Forrester also added that Indian organisations will move from AI experimentation to measurable outcomes as regulators and customers demand greater clarity on how AI systems operate.

It expects enterprises to face rising scrutiny on transparency and impact as AI becomes embedded in core business processes. In customer operations, Forrester predicts that three in ten enterprises will restructure teams to embed AI agents alongside human staff.

The firm said progress will depend on integrating these systems into legacy workflows and on managing tacit knowledge that AI tools cannot yet reliably capture.

Summarising the outlook, Ashutosh Sharma, VP and principal analyst at Forrester, said organisations would need to shift from early-stage AI enthusiasm to disciplined execution: “Leaders must move beyond the initial AI euphoria and embrace pragmatic innovation that drives business value. This means doubling down on governance, transparency, and measurable outcomes.”

The post ServiceNow May Acquire Boomi, A Fortune 500 Company to Sue a B2B Vendor: Forrester 2026 Predictions appeared first on Analytics India Magazine.

Indian and global IT workers are sounding alarms about the widespread misuse of Performance Improvement Plans (PIPs) to pressure employees to resign, often without a clear link to actual performance.

Workers’ accounts and HR leaders’ admissions reveal PIPs, meant for correction and development, are also used as opaque exit tools driven by metrics, staffing issues, and manager choices.

The anxiety has deepened after recent layoffs at TCS. It has encouraged employees across the industry to speak more openly about concerns they’ve had for years.

In a statement to AIM, TCS had said it is on a journey to become a future-ready organisation by implementing various strategic initiatives, including realigning its workforce model, and did not state that it was related to PIP.

Can of Worms?

A Pune-based developer, with over 10 years of experience and currently working at IBM, said the system is biased against workers from the moment they are placed on the bench or take permitted leave.

If you bench even without projects or take company-allocated leave, your utilisation will fall below 96.6% and result in a PIP. “There is no relation to your work or expertise,” he alleged.

The developer further said that the absence of a dedicated HR representative leaves workers at the mercy of managers whose decisions are effectively final.

On similar lines, a senior developer from Chennai, who previously worked at HCLTech, shared a similar experience, mentioning that he was placed on a PIP immediately after returning from a medical emergency.

Both developers preferred to remain anonymous, fearing repercussions for speaking out.

Meanwhile, IBM declined to comment, and HCLTech did not respond to queries from AIM about PIP policies.

Other IT companies remained unresponsive to our queries.

These individual accounts echo patterns that HR leaders say they have long observed.

Kaushik Kumar, founder of Kommunique Learning and a senior talent professional, said that PIP misuse is rarely an isolated HR failure but part of deeper structural and cultural problems within IT organisations.

He said misplaced incentives often push managers to adopt a survival mindset and prioritise control over development.

“In many IT companies, PIPs are now seen as an exit strategy rather than an improvement strategy,” Kumar said. “HR plays a procedural role and often lacks authority to challenge a manager’s decision, especially in offshore projects.”

An Aristo Legal commentary stresses that fairness requires clear expectations, documented performance gaps and a genuine chance to improve.

The Bengaluru-based legal firm said that Indian labour law lacks specific rules for performance-based termination, but Supreme Court precedents require that employees be informed of deficiencies, treated consistently and given a fair, well-documented opportunity to improve.

Kumar said a PIP cannot be initiated without an established record of feedback, adding that “surprise PIPs” are both procedurally invalid and can be challenged.

Meanwhile, Harpreet Singh Saluja, advocate at the Bombay High Court and president of the Nascent Information Technology Employees Senate, a body advocating for IT/ITES workers facing unfair practices like forced resignations, said the organisation has been receiving complaints about forced resignations disguised as “performance issues” in Indian IT.

He noted that under the Industrial Disputes Act, 1947, employees with at least one year of service are legally protected, and resignations given under pressure or coercion are invalid.

Saluja advised workers to refuse forced resignations, demand a written justification, preserve evidence, and seek intervention from the labour department if pressure persists.

He said HR teams often avoid formal documentation because coerced exits cannot withstand legal scrutiny.

While several HR leaders and legal experts warn against misuse, others argue that most IT companies follow robust processes.

TeamLease Digital CEO Neeti Sharma said most firms in India maintain transparent frameworks for KPIs, reviews and PIPs. “These processes include multiple rounds of feedback, and only then are employees put on PIPs,” she said.

With bench times shortening and niche skills in demand, Sharma said companies have strong incentives to ensure fairness rather than to pursue premature exits.

According to Unearthinsight, bench periods have now been reduced to 35–45 days, down from the 45–60 days seen in FY20 and FY21.

The shift, the firm said, is aimed at improving cost efficiency and resource utilisation.

Sharma said firms monitor patterns such as team-level attrition, bench cycles and exit feedback to identify unusual spikes.

“Sudden increases in PIPs are red flags for HR, prompting closer scrutiny of managers,” she said.

When employees raise concerns about unfair PIPs, Sharma said TeamLease’s clients (IT companies) engage managers to understand the rationale and then facilitate discussions involving both sides.

If the issue stems from skill mismatch, she said, redeployment to another project is considered to ensure the employee succeeds in a more suitable role.

Nandini Kantharaj, an experienced HR professional, said the broader issue is the mindset behind PIPs. “This mindset not only undermines the purpose of a PIP but creates fear and resistance among employees,” she said.

Kantharaj stressed the need for managers to identify gaps correctly, document expectations and coach employees consistently before escalating to formal plans.

Her concern mirrors many workers’ beliefs that PIPs are often just a procedural defense for pre-decided terminations.

She noted that unless organisations redesign performance systems to prioritise psychological safety, accountability and fairness over utilisation and optics, the misuse will persist.

The post How Indian IT Turned Performance Reviews into an Exit Tool appeared first on Analytics India Magazine.

When AWS introduced Kiro earlier this year, the company presented it not just as another AI assistant but as a new way to rethink how software should be developed.

At AWS re:Invent 2025 in Las Vegas, the cloud giant made no secret of its push to win over developers. With the launch of Kiro Powers and Frontier Agents, AWS says it is closer than ever to solving one of the oldest problems in software engineering by helping developers ship production-ready code faster, more reliably, and with less frustration.

The company said the agents are autonomous, scalable, and capable of operating for extended periods without intervention. It said the approach was shaped by three insights: that teams gain more value when agents pursue broader goals, run multiple tasks in parallel, and operate independently for long durations.

Frontier Agents

The launch of Frontier Agents includes the Kiro Autonomous Agent, the AWS Security Agent, and the AWS DevOps Agent.

In an exclusive interaction with AIM, Amit Patel, who leads engineering for Kiro, described 2025 as a period of discovery, rapid evolution, and unexpectedly strong customer demand.

Explaining why such agents are needed, Patel said that DevOps problems always strike at the worst possible moment.“These things happen at 2 o’clock in the morning,” he quipped.

Patel said that the DevOps Agent is built to prevent that, identifying incidents, analysing root causes, and even fixing them before teams are paged. The company said the agent has handled thousands of escalations internally, identifying root causes in an estimated 86% of cases.

On the other hand, the Kiro Autonomous Agent can plug into Jira or GitHub and pick up backlog items on its own. “Engineers can focus on building features,” Patel explained. Meanwhile, the agent can look at tickets and fix them.

Patel sees this as a breakthrough for reducing technical debt, one of the biggest productivity drains for engineering teams. The Security Agent can catch problems early, continuously check code, and remove the manual overhead of repeated audit cycles.

Spec Driven Kiro

Patel said that one of the most important lessons from early user tests of Kiro was that simple code completion, now a commodity feature across AI coding tools, was nowhere near enough.

That push led to one of Kiro’s defining capabilities, ‘spec-based development’. Developers can describe requirements in natural language, generate a design, break the work into tasks, and then have the system generate code, all within a structured workflow.

Patel described it as a way to preserve the fluidity of AI-assisted coding while forcing the system to think like an engineer rather than a text predictor.

Moreover, the new feature, Kiro Powers, gives AI agents extra skills whenever needed. It can pull in the right tools and knowledge on demand, such as Stripe, Figma, or Supabase integrations, by loading only the MCP tools and guidance required for the task.

Patel explained why this matters. “Inside Amazon, some teams load up 50 or 60 MCP servers… and you get a context problem. That leads to poorer results.”

Kiro Powers solves this by loading tools only when needed, keeping the context window clean. “It dynamically loads the relevant context at the relevant time,” Patel said. “It improves performance, reduces cost, and avoids context problems.” Postman is one of the early adopters of this tool.

Eventually, AWS wants Kiro Powers to be compatible with tools outside its platform and to adopt a more open-ecosystem approach than many competitors.

Adoption, Enterprise Needs, and India’s Role

Speaking about Kiro, Patel said that although enterprises tend to move slowly, interest is already surging. “It’s only been a couple of weeks since GA, but we’ve had a lot of enterprise interest,” he said.

Internal teams at AWS have become some of Kiro’s biggest users.

Moreover, Patel noted that enterprises are already asking for more robust governance controls. “One customer asked if they could have a Kiro Power specific to their enterprise, loaded on every installation and always used,” he said. “They don’t want deviations from coding patterns.”

Asked specifically about India, Patel said AWS isn’t segmenting capabilities by geography but expects strong adoption. “It’s going to be very interesting for India because we have such a big tech community,” he said. “Bangalore is the AI hub of India.”

Patel also spoke about how pricing models are likely to change. Kiro currently follows a seat-plus-credits structure, but background agents may require a different approach. “For asynchronous and cloud-based agents, you’ll likely see a usage-based model,” he said.

The post Meet AWS’ Frontier Agents Built to End Developers’ 2 AM Nightmares appeared first on Analytics India Magazine.

Replit and Google Cloud have expanded their strategic partnership to bring vibe coding capabilities to enterprise developers, marking a shift from a tool popular among individual programmers to one aimed at larger teams.

Under a new multi-year agreement, Replit will increase its use of Google Cloud infrastructure and services, further integrate Google’s AI models, and jointly support enterprise use cases through go-to-market efforts on Google Cloud Marketplace and co-sell programs.

Google Cloud will remain Replit’s primary cloud provider. Services such as Google Cloud Run, Google Kubernetes Engine, and BigQuery will continue to support Replit’s applications and future scaling needs. Replit’s platform already uses multiple Google Cloud services and integrates Gemini models through Vertex AI.

As part of the expanded collaboration, Replit now supports Google’s latest models, including Gemini 3, Gemini 2.5 Flash Lite, Gemini 2.5 Flash, and Imagen 4 to power coding and multimodal features. Replit recently rolled out Gemini 3 in its Design mode.

Meanwhile, Google also launched Antigravity, its new agentic AI IDE embedded with Gemini 3 models.

Thomas Kurian, CEO of Google Cloud, said the expanded partnership will deliver more capabilities to Replit’s users through deeper integrations with AI and cloud services and will accelerate the adoption of vibe coding in the enterprise.

Replit co-founder and CEO Amjad Masad said the company is seeing rising interest from large organisations. “Over the last few months, we have seen fantastic adoption in businesses, especially in the Fortune 1000,” he said.

“Today’s expanded partnership with Google will enable us to scale faster and more deeply as we integrate Google’s offerings with ours — the work is just beginning.”

The companies said they aim to help enterprise developers build applications through conversational interfaces, a workflow that has gained traction this year as AI-powered coding tools become more widely used.

The post Replit Brings Vibe Coding to Google Cloud for Enterprises appeared first on Analytics India Magazine.

The US Department of State has announced that beginning December 15, all H-1B visa applicants and their H-4 dependents will undergo a mandatory online presence review, extending a screening requirement that previously applied only to international students and exchange visitors.

To support this expansion, the department has directed applicants across the H-1B, H-4, F, M and J non-immigrant visa categories to set the privacy settings of all social-media accounts to “public.”

According to the announcement, the state department relies on “all available information” to identify visa applicants who may be inadmissible to the United States, including individuals who could pose threats to national security or public safety.

The department also said it already conducts online-presence reviews for F, M and J applicants and is now bringing H-1B and H-4 applicants under the same process.

Reiterating that “every visa adjudication is a national security decision,” the department emphasised that consular officers must confirm applicants do not intend to harm US interests and can credibly demonstrate eligibility for the visa category they are seeking.

It added that holding a US visa is “a privilege, not a right.”

Until now, the mandatory online-presence review applied only to F-1 and M-1 students and J-1 exchange visitors, categories that have been subject to enhanced digital-screening measures for several years.

For H-1B workers and H-4 dependents, consular officers could review publicly available online information at their discretion, but there was no universal, formal requirement.

Applicants were not instructed to make their social-media accounts public, and online-presence checks were not a standardised part of the adjudication process.

The US Citizenship and Immigration Services data shows that India overwhelmingly dominates H-1B approvals, accounting for 71% (283,397) of all 399,395 approved beneficiaries in FY 2024.

China follows at 11.7%, with other countries contributing only small single-digit shares.

Indian IT companies that were one of the largest beneficiaries of H-1B visas have been cutting down significantly on applications over the past few years.

Total petitions from eight leading firms, including Tata Consultancy Services, Infosys, and Wipro dropped from 25,475 in 2022 to 14,319 in 2025, according to US Citizenship and Immigration Services.

The figures indicate a 44% drop in visa petitions by Indian IT firms over a four year period.

The post From Dec 15, Your Social Media is Part of H-1B/H-4 Screening appeared first on Analytics India Magazine.

The Karnataka government has signed an MoU with Taiwan-based Allegiance International to establish an India–Taiwan Industrial Technology Innovation Park (ITIP) to bolster the state’s electronics and semiconductor manufacturing ecosystem.

Under the agreement, the Allegiance Group will invest ₹1,000 crore over the next five years to develop the park, which is expected to generate around 800 direct jobs as Taiwanese firms set up advanced manufacturing, chip design, and R&D units.

Officials said the ITIP will operate as a dedicated hub for Taiwan’s electronics supply chain players, enabling technology transfer, component manufacturing, PCB production, and design innovation. The state expects the project to significantly strengthen Karnataka’s position as a leader in the Electronics System Design and Manufacturing (ESDM) sector.

CM Siddaramaiah said the collaboration “marks a significant step towards making Karnataka a global hub for semiconductor innovation,” adding that it will expand opportunities for local talent and bolster India’s position in the global value chain.

IT & BT Minister Priyank Kharge said the partnership reinforces Karnataka’s dominance in deep-tech and semiconductor design, noting that the government is committed to creating a level playing field for global investors in advanced technologies. Allegiance Group vice-president Lawrence Chen said the ITIP will “serve as a catalyst for Taiwanese companies looking to scale in India.”

The MoU aligns with national efforts under Make in India, Atmanirbhar Bharat, and the India Semiconductor Mission.

This push toward high-tech manufacturing comes after Burkhan World Investment (BWI) announced a ₹1,500-crore plan to build a GPU and AI server manufacturing facility near Devanahalli on Bengaluru’s outskirts.

Backed by BWI subsidiary The Ghazi Group (TGG), the project aims to produce processors and computing systems for next-generation AI workloads. The state government said the project will bring together global and Indian partners — including Gigabyte, Sera Network, Resolute Group and Quad-Gen — to build an end-to-end AI hardware value chain, from components to export-ready assembly lines.

The post Taiwan’s Allegiance Group To Invest ₹1,000 Cr In Karnataka’s Semiconductor Push appeared first on Analytics India Magazine.