Priyanka Tiwari
Share
Suspense crime, Digital Desk : The new features added by Microsoft and Google in May 2025 further enhance the use of AI in Google’s AI Mode and Microsoft’s autonomous GitHub Copilot. AI now aids users in every step from searching and coding, to even collaboration. Tools like these provide capabilities such as automated research, real time translation, and even interactive AI agents that assist every step of the way.
Microsoft’s Discovery platform and Google’s Deep Search enable enterprise-level workflows that automate complex tasks. Gemini 2.5 Pro and GPT-4o, newer models powered by Nvidia alongside their hardware, enable these developments. These AI systems are not just efficient—they are revolutionary in scale and reach.
First-rate AI solutions tend to cost a lot. The adoption cap from Google’s AI Ultra subscription, for $249.99 per month, demonstrates this charge. Premium features like Gemini 2.5 Pro and Deep Think Mode are likely to remain under paywalls. Furthermore, Microsoft’s promising Azure and Windows AI Foundry rest atop a pile of expensive infrastructure, putting small businesses at a disadvantage with no end to innovation in sight.
The self reliant AI tools offer even more expensive features, like with Microsoft’s multi-agent planners and Google’s Project Mariner. These place a growing financial burden on earth, showcasing disruption to the environment.
The Growing Issue of the Environmental Impact of AI Technology
AI's energy consumption is hitting an unsustainable level. For example, services such as Google Live Search, or even on-device models like Microsoft's Phi 4 mini, rely on extensive grid-like networks of servers. Each month, Google’s AI Overviews alone supports more than 1.5 billion users, which is further straining the energy resources across the globe.
It is said that the data centers which power Gemini 2.5 or GPT-4o models use upwards of 25 terawatt-hours a year, nearly equating the energy demands of several million homes. If advances in renewable energy technologies are not adopted rapidly, AI will severely hinder the already slow global climate initiatives.
Ethics and Responsibility in AI Technology: Privacy, Accountability, and Bias
The rollout of new capabilities of AI technologies is already stirring privacy and ethical issues. Microsoft’s NLWeb framework, that gives AI access to proprietary data locked away in private web pages, poses concerns of classified data being exposed if managed poorly. Other worries include Google’s virtualTryOn tool and AI Mode, which utilize a person’s image data and search history.
Blurring the lines of responsibility further, agentic AI systems make questions of liability vague: if Gemini makes a booking or Copilot writes harmful code, who takes the blame? This is principally exacerbated in AI fueled decision making systems and algorithms—especially in sensitive industry sectors like healthcare or finance—with AI’s opaque the decision making sinews of. Most importantly, fairness is still at risk due to bias, like in the case of AI-related drug discovery or in recruitment using in-built evaluation systems, which is more likely to be uncompromisingly biased due to training data.
Regulatory Frameworks Are Struggling To Keep Up
The existing regulations do not capture the full extent of AI’s impact. While some regions like the EU and Malaysia have introduced rules like energy reporting or bias audits, most countries have no implementable benchmarks at all. Tools like Deep Search and the Discovery platform work in unregulated spheres that could lead to the monopolization of power and stifling of competition.
Furthermore, the cross-border nature of AI operations poses challenges to legal definitiveness. AI systems are trained on a global dataset, run on an international cloud, and interact with users worldwide. This creates gaps in jurisdiction and enforcement.
Policy Directions Towards Safer AI
There is no question that the government must intervene. The key ones that need involvement are:
Transparent Report: The disclosure of AI energy usage should be made public.
Sustainable Policy Framework: AI tax incentives should be tied to renewables and non-polluting forms of energy.
Broadband Access: Provide subsidized AI tools to public institutions and small businesses.
Bias and Safety Audits: Mandate external scrutiny for the most impactful AI applications.
Legal Definition: Determine who is accountable for autonomous AI decisions in healthcare or finance.
These measures are steps in the right direction to encourage responsible innovation without stifling developers and companies.
Global Collaboration is the Path Forward
To achieve the most important goals of AI, it is critical to achieve the cooperation of all the regions of the world. There is a need to adopt an international approach towards the development of AI. Everyone in the world needs to work together for the development of its technology. Society must foster open-source initiatives, governments spend resources to create a clean AI framework, and global entities put together ethics policies for all nations to follow.
AI's capability to aid research and productivity is evident in products like Gemini's Agent Mode and Microsoft's Copilot. However, such inventions must also adhere to human rights and the principles of access equity, environmental sustainability, and socio-economics.
Read More: New Tata Sierra Sets National Mileage Record With Hyperion Engine
