The artificial intelligence landscape is rapidly evolving, marked by significant investments, new capabilities, and emerging threats. Major tech companies like Oracle are transforming into AI powerhouses, leveraging cloud infrastructure and Nvidia GPUs to secure massive deals with AI firms including Meta, XAI, and OpenAI, reporting a substantial backlog of $455 billion. Meanwhile, the investment boom in AI infrastructure is substantial, with global spending on data centers projected to exceed $3 trillion by 2028, though questions about the ultimate return on these nearly $400 billion annual investments persist. New forms of AI, such as agentic AI, are moving beyond intelligence to take action, enabling complex task planning and execution, with companies like Thomson Reuters integrating these capabilities into legal and tax products. However, this rapid development is not without its risks. Security experts warn that attackers are exploiting AI integration before adequate security measures are in place, with AI-driven attacks moving over 40 times faster than traditional methods and bypassing identity defenses through compromised credentials. AI-powered cloud security platforms are becoming essential to combat these threats, automating monitoring and reducing incident response times. Amidst these advancements, some researchers, like Eliezer Yudkowsky, founder of the Machine Intelligence Research Institute, are issuing stark warnings about the existential risks posed by superintelligence, advocating for a complete halt to AI development. In education, initiatives like Cognizant's 'AI Pledge to America's Youth' aim to boost AI literacy by training teachers and students, while law schools grapple with how to train future lawyers to use AI tools effectively, with some suggesting a return to oral assessments. Accenture's CEO also highlights AI as a key driver of long-term growth, with the company training its large workforce in agentic AI.
Key Takeaways
- Oracle is experiencing significant growth in the AI era, driven by its cloud infrastructure and Nvidia GPU offerings, securing billion-dollar deals with firms like Meta, XAI, and OpenAI, and reporting a $455 billion backlog.
- Global investment in AI infrastructure is immense, with nearly $400 billion spent annually and projected data center spending to exceed $3 trillion by 2028, raising concerns about investor returns.
- Agentic AI represents a new phase, enabling AI to plan and execute complex, multi-step tasks, with applications emerging in fields like legal and tax services.
- Cybersecurity threats are escalating due to AI, with attacks moving significantly faster and bypassing traditional defenses by exploiting stolen credentials, necessitating AI-powered security platforms.
- Concerns about existential risk from superintelligence are being raised by prominent AI researchers, who advocate for halting AI development due to unpredictable outcomes.
- Cognizant is launching an initiative to enhance AI literacy among US teachers and students, aiming to train tens of thousands by 2028.
- Law schools are facing challenges in preparing students for an AI-driven legal field, with discussions around teaching AI tool usage and adapting assessment methods.
- Accenture's CEO identifies AI as a primary driver of long-term growth, with the company actively training its workforce in agentic AI.
- AI is improving accuracy and profitability in industries like construction bidding, with companies achieving high win-loss forecasting accuracy through machine learning models.
- Human error accounts for over 55% of cloud breaches, highlighting the need for AI-driven cloud security platforms to automate monitoring and improve threat detection.
CISOs face new AI threats and security gaps
A SANS Institute report warns that attackers are exploiting the rapid integration of AI into business processes before security measures are in place. AI-driven attacks can move over 40 times faster than traditional methods, making breaches happen before alerts are even seen. Many security operations centers (SOCs) are using AI tools without custom security plans, leaving them vulnerable to threats like prompt injection. To combat this, experts recommend an adoption-led control plane that provides security and efficiency, focusing on clear outcomes rather than abstract benchmarks. The report also outlines a framework to Protect, Utilize, and Govern AI to close these security gaps.
AI platforms to transform cloud security in 2025
Cloud-first operations have expanded enterprise attack surfaces, making traditional security tools insufficient. AI-powered cloud security platforms are becoming essential in 2025 to provide faster, smarter defenses. Human error causes over 55% of cloud breaches, and AI platforms help by automating monitoring, detecting anomalies in real-time, and learning from new attack patterns. These platforms also reduce incident response times and improve threat accuracy. With the average cloud breach costing $4.35 million and a shortage of cybersecurity talent, AI-driven solutions are crucial for proactive defense and compliance.
AI attacks bypass security as hackers log in with stolen credentials
Generative AI is accelerating cyber threats, with voice phishing and deepfakes seeing significant growth. A recent report shows that 79% of AI-driven attacks now bypass traditional identity defenses by simply logging in with compromised credentials. Attackers use AI to create and abuse identities at scale, moving laterally within networks in just minutes. Traditional identity and access management systems struggle to keep pace with these fast-moving threats. Companies like Cushman & Wakefield are adopting AI-powered platforms that use behavioral intelligence to monitor identities and take real-time action against anomalies, securing access across cloud, endpoint, and application domains.
AI doomsayer warns of existential risk from superintelligence
Eliezer Yudkowsky, founder of the Machine Intelligence Research Institute, warns that building advanced AI systems poses an existential threat to humanity. In his new book, 'If Anyone Builds It, Everyone Dies,' co-written with Nate Soares, he argues that any superintelligence created with current techniques will inevitably lead to human extinction. Yudkowsky, who influenced early AI leaders like Sam Altman and Elon Musk, believes intelligence and benevolence are separate traits, and a superintelligent AI could harm humans as an unintended consequence of its goals. He advocates for stopping AI development entirely to prevent global disaster.
AI researchers warn of global catastrophe from rapid development
AI researchers Eliezer Yudkowsky and Nate Soares, authors of 'If Anyone Builds it, Everyone Dies,' caution that the rapid development of artificial intelligence by major tech companies could lead to human extinction. They argue that current AI models are poorly understood and could eventually surpass human control, using Earth's resources for their own purposes. The authors believe that even slowing down development or building alternative systems is not enough, and that all AI development should cease. While some dismiss their warnings as alarmist, they emphasize the unpredictable nature of advanced AI and the potential for catastrophic outcomes.
Oracle transforms into an AI powerhouse with strong cloud growth
Once considered a dot-com relic, Oracle is now a significant player in the AI revolution, driven by its cloud services. Oracle Cloud Infrastructure competes with AWS and Azure, offering a full-stack approach by providing Nvidia GPUs and AI models to customers. The company recently reported a massive backlog of $455 billion in Remaining Performance Obligations, a 359% year-over-year increase, and signed billion-dollar deals with major AI firms like XAI, META, NVDA, and OpenAI. Oracle's multi-cloud business also saw astonishing growth of 1,529%. This strong performance has shifted investor perception, positioning Oracle as a key growth engine in the AI era.
AI improves construction bidding accuracy and profitability
Western Specialty Contractors is using artificial intelligence to improve construction bidding accuracy and profitability. By analyzing years of bid history with a custom machine learning model, they achieved over 70 percent accuracy in win-loss forecasting, allowing them to prioritize opportunities and allocate resources effectively. This AI integration helps address challenges like rising costs and labor shortages in the construction industry. While many firms use AI applications, they often struggle with fragmented data and paper-based processes. Western emphasizes that clean, consistent data and embedding AI insights into daily workflows are crucial for successful AI adoption.
Agentic AI moves beyond intelligence to take action
Agentic AI represents a new chapter in artificial intelligence, capable of not just responding but also acting on complex objectives. Unlike traditional AI interfaces, agentic AI can plan multistep tasks, adapt in real-time, and execute workflows alongside professionals. Companies like Thomson Reuters are integrating these capabilities into products like CoCounsel Tax and CoCounsel Legal, allowing users to delegate complete assignments such as tax returns or legal motions. This evolution requires platforms that prioritize trust, accuracy, and domain expertise, with AI engineers working alongside human experts to ensure reliable and aligned outcomes.
AI investment boom faces uncertain payback
The current boom in artificial intelligence investment, with companies spending nearly $400 billion on AI infrastructure this year, is one of the largest in modern history. Major AI model makers like OpenAI and Anthropic are raising billions, approaching a combined valuation of half a trillion dollars. Analysts predict global spending on data centers for AI will exceed $3 trillion by the end of 2028. Given the immense scale of these investments, it raises questions about the eventual returns, suggesting that even if the technology succeeds, many investors may lose money, and a failure could lead to severe economic and financial consequences.
Cognizant pledges AI training for US teachers and students
Cognizant Technology Solutions Corporation is supporting a national initiative to expand AI education in the United States. As part of its 'AI Pledge to America's Youth,' the company committed philanthropic funding, educational resources, and employee volunteering to promote AI literacy. By 2028, Cognizant aims to train thousands of teachers and tens of thousands of students to prepare them for an AI-driven economy. This initiative also complements Cognizant's global upskilling program, Synapse, which aims to train one million people worldwide by the end of 2026 to help them adapt to digital transformation.
Law schools grapple with training lawyers in the AI era
Training young lawyers in the age of AI presents a significant challenge for law schools. Economist Tyler Cowen suggests that law schools must teach students how to effectively use AI tools. Academician Clay Shirky argues that educational institutions should shift from take-home assignments to in-class assessments like oral exams to ensure students are truly learning and not just using AI to cheat. Traditional methods like the Socratic method, which emphasizes oral questioning, could be ideal for this new era, but many law professors have moved away from it. Law firms also need to invest more in mentorship and hands-on training that AI cannot fully replace.
Accenture CEO: AI will drive long-term growth
Accenture CEO Julie Sweet believes artificial intelligence will be a primary driver of long-term growth for the company. Speaking on 'Bloomberg Tech,' Sweet stated that Accenture is actively training over 700,000 employees in agentic AI. This focus on AI is part of the company's strategy to adapt and lead in the evolving technological landscape, ensuring its workforce is equipped for future advancements.
Sources
- CISOs brace for a new kind of AI chaos
- AI-Powered Cloud Security Platforms In 2025
- They don't break in, they log in: 79% of gen AI attacks now bypass legacy identity defenses
- A.I.’s Prophet of Doom Wants to Shut It All Down
- Researchers give doomsday warning about building AI too fast
- Oracle (ORCL): From Dot-Com Relic to AI Powerhouse
- Smart Bidding Strategies with Artificial Intelligence
- For Agentic AI, Intelligence Is Only Half The Battle
- What if the $3trn AI investment boom goes wrong?
- Cognizant pledges AI training for teachers, students in the US
- Training Young Lawyers In The Age Of AI: A Riddle Wrapped In A Mystery Inside An Enigma
- Accenture CEO Sees AI as a Long-Term Growth Driver
Comments
Please log in to post a comment.