Business & Ethical AI

Pieces on AI’s impact on business processes, ROI, leadership decisions, plus the risks, ethics, and reliability of these technologies.

379 articles • Page 4 of 26

Jul 16, 2026

Businesses Prepare for 2026 AI Agent Liability with New Oversight Rules

Businesses may face new liability rules for AI agents in 2026 and should prepare by keeping clear, documented oversight of their autonomous software. The law still treats AI as a tool, so organizations remain responsible unless they can show they monitored and controlled their systems. Suggested steps include having contracts that match different risk levels, keeping detailed logs and audit trails, and setting up quick-response teams for handling incidents. Some systems, like the Internet Court, might help settle disputes quickly, but their authority is limited for now. Updating processes and assigning clear human roles for each AI action may help companies meet these changing legal expectations.

Jul 15, 2026

xAI's Grok Build CLI Uploads Entire Repos, Secrets to Google Cloud

Security researchers say xAI's Grok Build CLI may upload entire code repositories, including private files and secrets, to a Google Cloud bucket by default. Reports suggest this happened even if the privacy setting was turned off, and included files that were supposed to be excluded. xAI appears to have stopped these uploads after the issue was made public, but it is unclear how much data was exposed or if it was deleted. Researchers recommend developers change any exposed secrets and check new settings to prevent uploads. Experts suggest this case shows why all new AI tools should be checked for similar security risks.

Jul 14, 2026

Irys Unveils Trust-First Legal AI After Public Live Demo

Irys showed its new legal AI tool in a live demo, letting lawyers see how it works and why trust and transparency are important. The system may help lawyers by showing its sources and reasoning, making it easier to check its work. Early users report the tool saves time and increases efficiency, but not all lawyers fully trust AI yet. New rules in the EU soon may require AI tools to be more open about how they work. Irys's demo suggests the company is focusing on building trust to encourage more lawyers to use their AI.

Jul 13, 2026

Startups: Contract Strategies Help Manage Vendor & Supplier Risks

Managing supplier and vendor risk has become essential for startups and their partners, especially as financial pressures may affect many AI startups by 2027. Using retention of title clauses and clearly identifying goods may help suppliers protect their interests if buyers cannot pay. Experts suggest splitting payments into milestones and running thorough checks on startups before signing contracts, such as verifying legal names and financial health. Mistakes in paperwork or unclear contracts might weaken a supplier's ability to recover assets. Regular monitoring, diversifying clients, and having backup plans appear to be important strategies to manage these risks.

Jul 12, 2026

New AI Agent Checklist Validates Enterprise AI, Cuts Project Risk

Enterprises may need a clear audit checklist to tell real AI agents apart from simple chatbots, as more rules and claims of 'agentwashing' appear. Analysts suggest that over 40 percent of AI agent projects might be canceled by 2027 if buyers can't trust what vendors offer. The checklist should test for things like clear tracking of all agents, strong controls to stop agents from doing too much, quick ways to stop risky actions, and secure audit records. It also appears important to have human checks for high-risk decisions and to make sure agents can recover from problems without making many mistakes. These steps may help companies pick trusted AI agents and lower the risk of project failure.

Jul 12, 2026

LexisNexis Reports Misinformation as Top AI Risk for Consultants

A recent LexisNexis report suggests that about half of consultants see misinformation as the biggest risk when using AI. Consultants may spend extra time checking AI results, since mistakes can happen in different ways, like factual errors, outdated data, model misuse, or bias. Firms often use a mix of automated tools and human checks to catch and fix these problems, but no single solution works for everything. The report notes that most clients manually verify AI outputs and expect clear plans for fixing mistakes. It also appears that many consultants use unapproved AI tools, so strong rules and regular checks are needed to lower risks and learn from errors.

Jul 12, 2026

AI Hallucinations Cost Companies Billions, Force New Governance in 2026

AI-generated mistakes, called hallucinations, have reportedly cost companies a lot of money and trust, with estimated global losses at $67.4 billion. Analysts say these mistakes may happen in 15 - 25 percent of routine financial tasks, and one incident in 2026 appears to have lost $2.3 billion because of fake data. Experts suggest the real cost includes both direct losses and the extra time employees spend checking AI work. New rules for 2026 recommend making sure humans check important AI decisions, using risk ratings, and keeping records for review. Training teams to spot common AI errors and requiring proof for claims may help catch problems early and protect companies.

Jul 12, 2026

US companies adopt China's GLM-5.2 AI, raising security concerns

Some U.S. companies are using the Chinese AI model GLM-5.2 because it may be much cheaper than American models. Officials warn that using this technology might put important data and supply chains at risk, and there are questions about long-term security. Experts say GLM-5.2 could be used for hacking and may fall under Chinese laws if run on Z.ai's cloud. There is no clear federal ban yet, but some government officials and analysts suggest new rules could be needed. The main debate is whether the cost savings are worth the possible dangers.

Jul 12, 2026

OpenAI, Google Supplied AI Models to Pentagon-Blacklisted Chinese Firms

This situation suggests there are loopholes in export controls that let Chinese firms access US technology.

Jul 12, 2026

Consulting Firms Adopt Policies to Combat Shadow AI Data Leaks

Consulting firms are making it a top priority to require approvals for using AI tools because shadow AI may be spreading faster than rules can keep up. Reports suggest that many workers, especially in professional services, might be sharing sensitive data with unsanctioned AI platforms, raising the risk of leaks and breaking confidentiality agreements. Short policies that list approved tools, require human checks, and set clear data rules may help prevent these problems and meet new regulations. Experts say technical controls and regular training could be necessary, since many AI tools operate without managers knowing. Audits, logging, and fast reporting of any issues appear to be important for keeping data safe.

Jul 12, 2026

Consulting Firms Adopt Playbook to Validate Client-Facing AI Workflows

Consulting firms are starting to use a playbook to check and approve their AI work for clients, which may help keep quality high even when deadlines are tight. The process includes quick checkpoints at several steps, with different team members reviewing and signing off on the work. Tools that track sources, changes, and approvals seem to help make these reviews easier. Key performance indicators, such as how often work is finished on time or by AI, may show if the process works. Some reports suggest that tracking every AI prompt and having regular reviews might reduce unapproved AI use.

Jul 12, 2026

Unilever Expands Influencer Program to 300,000 Creators, Boosts Ad Spend

Unilever may grow its influencer program to 300,000 creators by the end of 2025, moving about half of its ad budget into social content. The company says it uses AI tools to help with finding and managing creators, but people still make creative decisions and keep relationships strong. Early signs suggest this shift might have helped brands like Dove grow, and Unilever plans to try new ideas like sharing sales with creators in some countries. The company appears to believe that keeping some decisions in human hands protects authenticity as it works with more creators. Future plans include stronger tracking, content credentials, and more ways to work with creators across all its main brands.

Jul 7, 2026

Amnesty: Generative AI Is 'Unlawful by Design' Due to Web Scraping

Amnesty International's report suggests that generative AI systems may violate human rights because they often use data scraped from the web without consent. Evidence in the report links these practices to possible privacy violations and discrimination. The report calls for governments to ban AI models built on illegally scraped data and for companies to stop collecting personal information without permission. Lawmakers and regulators in the EU and US appear to be considering new rules or bans based on these findings, but discussions are still ongoing. The outcome of these talks may lead to stricter rules for AI companies in the future.

Jul 5, 2026

Meta Curbs Employee AI Use After Projecting $145 Billion in 2026 Spending

Meta is limiting how much its employees can use AI after seeing forecasts that it may spend up to $145 billion on AI by 2026. This move appears to balance big infrastructure spending with daily costs from many employees using AI tools. Some early studies suggest that these limits might reduce wasteful behavior but could also lower innovation if employees feel restricted. Experts say that usage caps work best when combined with clear rules about who can use which tools. If not managed well, such limits may simply push employees to use personal accounts and reduce the benefits of earlier investments.

Jul 4, 2026

Enterprises Adopt New Playbooks to Manage Foreign AI Model Risks

Enterprises are making new plans to handle the risks of using foreign AI models, which may include security, legal, and compliance problems. Experts suggest that issues like hidden data tracking, cross-border data sharing, and copying models raise many concerns. Companies are now using guides and frameworks to check and track all outside AI models, and are updating contracts and technical controls to better manage these risks. Reports suggest that watching for unusual network activity and strong vendor rules may help spot and stop problems early. It appears that combining technical, legal, and contract steps is becoming important for strong AI risk management.