Google’s Gemini 2.5 “Computer Use” model can now interact with web pages visually — clicking, typing, dragging — just like a human using a browser.

AI agents in enterprise: AWS rolled out enhancements to its Strands Agents SDK, making it easier for developers to compose multiple autonomous agents that coordinate.

Intuit’s push: The company now embeds AI agents into its Enterprise Suite for automating accounting, payments, customer support, and more.

Emerging players: A startup named Reflection created an agent called Asimov, which reads code, emails, Slack threads, and documentation to infer project context.

Agents are vulnerable. Recent research shows widespread risk of hijacking attacks, where bad actors manipulate agent behavior or data flows.

“Agent washing” is becoming a concern — vendors overclaiming “agentic AI” to ride hype waves.

Security and identity for agents is complex. Without proper controls, agents may end up overprivileged or misused.

Productivity amplification: Workers will delegate repetitive, tedious, data-heavy tasks (e.g. report generation, inbox triage, data reconciliation) to agents, freeing humans for strategic, creative work.

New hybrid roles: Job titles like “Agent Supervisor,” “AI Prompt Engineer,” or “Ethical AI Auditor” will emerge. Workers will need both domain knowledge and AI fluency.

Skill premium shifts: According to the Future of Jobs Report 2025, AI and information processing rank among the top technologies expected to transform business.

Workforce strategy rewrites: Organizations will pivot to “frontier firms” — blurring lines between human judgment and autonomous AI systems.

Trust and oversight: As AI agents take more initiative, verifying their decisions becomes critical — “trust, but verify.”

Job displacement anxieties: In markets like the U.S., reports warn of layoffs and slow hiring attributed to AI adoption.

Unionization and worker power: In response to automation pressures and job instability, white-collar unionization is increasing.

Mental health & burnout: The burden of supervising AI systems — chasing anomalies, caught in “agent errors” — may add hidden stress.

Develop AI oversight skills: Think like a manager — delegate to agents, but routinely audit, question anomalies, and adjust.

Focus on human-only strengths: Emotional intelligence, critical thinking, ethical judgment, cross-disciplinary collaboration will matter more than ever.

Continuous learning becomes nonnegotiable: Reskilling, upskilling, staying curious — this is the new career baseline.

Establish guardrails: Companies must set clear policies on agent access, transparency, identity, and fallback protocols.

$2.8B: Total global creator earnings projected for 2025 Q3

78%: Of creators say “building a newsletter” is key to platform independence

51%: Of Gen Z prefers following creators over traditional celebrities