How AI Can Improve Investment Decisions

What if your portfolio manager never slept, never panicked, and processed every news headline on the planet before your morning coffee?

That is no longer a fantasy. Artificial intelligence is fundamentally changing how people invest — and the pace of that change is accelerating faster than most investors realize. Whether you are a retail trader trying to make sense of volatile markets or a seasoned fund manager looking for an edge, understanding how AI investment decisions work has gone from a competitive advantage to a survival skill.

In 2026, 59% of investors hold AI-related stocks according to a Motley Fool survey, and the global AI market — currently valued at $371 billion — is projected to reach $2.4 trillion by 2032. The numbers alone tell you this is not a trend to ignore. But AI is not just an investment theme; it is also the engine quietly powering the investment decisions being made around you, right now.

This article breaks down exactly how AI improves investment decisions — from predicting stock movements and executing algorithmic trades, to managing portfolio risk and eliminating emotional bias. You will also discover the tools used by institutions and individuals alike, the real limitations you need to know before trusting any algorithm, and where the technology is headed next.

1. How AI Is Reshaping Investment Decisions: A Big Picture View

For most of market history, investment decisions were driven by a combination of human intuition, spreadsheet analysis, and whatever information a portfolio manager could digest before the trading day opened. That model has a ceiling — and the market figured out its limits long ago.

AI does not simply automate those old processes. It introduces entirely new capabilities. Machine learning models can simultaneously analyze thousands of variables — earnings data, central bank signals, geopolitical risk, even satellite imagery of retail parking lots — and update their conclusions in real time. Natural language processing (NLP) reads earnings call transcripts within seconds of release, scoring the sentiment of every phrase the CFO uses. Deep learning models identify non-linear patterns across decades of historical price data that no human analyst would ever spot.

The scale of institutional commitment to this shift makes the stakes clear. McKinsey estimates that AI could add $3.8 trillion in annual profits to the global financial services industry. Morgan Stanley Research reports that 21% of S&P 500 companies mentioned at least one AI-related benefit in 2025 — double the rate from 2024 alone. And according to the OECD, venture capital investment in AI firms globally surpassed $225.8 billion in 2025, nearly double the prior year’s total.

AI is not replacing investment judgment. It is expanding the scope of what is possible to judge — processing information no human team could ever match in speed or volume.

If you want to understand which AI analytics platforms are driving these decisions across business and finance contexts, the guide on the best AI analytics tools for small businesses in 2026 offers a solid starting point for understanding the analytical stack.

2. AI in Stock Market Prediction: How Accurate Are the Algorithms?

Predicting where a stock price is heading is one of the oldest — and hardest — problems in finance. AI has not solved it perfectly. But it has materially improved the odds.

In 2025, advanced machine learning models improved forecasting accuracy from approximately 80% to 90% in controlled backtesting environments, according to research published by Qubit Capital. That ten-point gap may sound modest, but at institutional scale, it translates to hundreds of millions of dollars in improved trade outcomes.

The key advantage AI brings to prediction is not smarter guesses — it is the sheer breadth of what gets analyzed. Traditional models looked at a few dozen financial variables. Modern AI systems track hundreds simultaneously, layering in alternative data sources like consumer spending patterns, social media sentiment, shipping container volumes, and even the tone of central bank press releases.

NLP and Sentiment Analysis: Reading the Market’s Mood

Natural language processing has become one of the most powerful prediction tools available. AI systems now read and score the sentiment of corporate earnings call transcripts within seconds of their release — flagging whether a CEO sounds confident or evasive, whether guidance language has shifted toward caution, or whether the word choice in a press release has changed from last quarter.

Sentiment analysis tools also scan news headlines, Reddit threads, financial blogs, and Twitter/X activity to detect early signals of shifting market opinion. The speed at which these systems operate gives institutional users a measurable edge over manual analysts — what once took an analyst team two days to synthesize now takes seconds.

A real example from early 2025: when DeepSeek launched and disrupted the AI narrative, AI-powered sentiment systems flagged extreme negative momentum in semiconductor stocks before much of the retail market had processed the news. Nvidia’s stock dropped 17% in a single day — a move that predictive AI models had flagged as a high-probability risk scenario.

Multimodal AI: Combining Every Signal Into One View

The newest generation of market prediction tools goes beyond text. Multimodal AI combines data from audio sources (earning call recordings), financial filings, on-chain blockchain metrics, and real-time news feeds to produce a unified, holistic market signal. According to LiquidityFinder, these platforms improve trading signal accuracy by leveraging the full range of available information — not just what appears in a price chart.

3. Algorithmic Trading Powered by AI: Speed, Precision, and Scale

Algorithmic trading is not new. Rules-based automated systems have been executing trades since the 1980s. What is new is the intelligence behind those rules. AI-powered algorithmic trading does not just follow if-then logic — it adapts, learns, and optimizes continuously.

These systems execute trades in microseconds — far faster than any human can process information, let alone act on it. The result is increased market efficiency, narrower bid-ask spreads, and better liquidity for all participants. Goldman Sachs, Renaissance Technologies, and Two Sigma have been running sophisticated AI-driven quantitative trading operations for years, and their results have consistently beaten traditional actively managed approaches over time.

Deep reinforcement learning has become a particularly powerful technique in this space. Unlike supervised learning models that train on historical outcomes, reinforcement learning agents learn by interacting with simulated markets — discovering optimal trading strategies through trial and error across millions of simulated scenarios before they touch a single real dollar.

The Flash Crash Risk: When Algorithms Go Wrong

It would be intellectually dishonest to discuss AI trading without addressing the systemic risks. In 2025, a widely documented algorithm malfunction caused a sudden 6% drop in the S&P 500 within minutes before circuit breakers halted trading. The event was a reminder that when multiple AI systems are trained on similar data and trade the same signals, they can amplify volatility rather than reduce it.

The debate about whether AI increases or decreases overall market stability remains genuinely open. What is clear is that AI trading operates in a regime where correlated behavior among many systems is a real systemic risk — one that regulators at the SEC and ESMA are actively working to understand and constrain.

Quantum Computing: The Next Frontier in Algorithmic Strategy

On the horizon sits a technology that could make today’s AI trading look slow by comparison. Goldman Sachs’ Quantum Studio has already demonstrated the ability to minimize bond portfolio risk by up to 40% while solving complex optimization problems 100 times faster than classical computers. As quantum-AI hybrid systems mature, the advantage available to institutions that deploy them will be substantial.

4. AI-Powered Risk Assessment: Smarter Protection for Your Portfolio

Risk management is where the rubber meets the road for most investors. Making money is satisfying; avoiding catastrophic losses is essential. AI has significantly advanced the science of identifying, measuring, and mitigating investment risk.

Traditional risk models typically analyzed market risk in isolation — how much might this position lose if the market falls? Advanced AI systems simultaneously analyze market risk, credit risk, liquidity risk, and operational risk in one integrated model. They run continuous stress tests, model tail risk scenarios, and flag exposure concentrations that a portfolio manager reviewing a spreadsheet would miss.

A concrete example: Goldman Sachs has deployed an AI risk model internally known as Risk Guardian, which successfully predicted the potential impact of specific geopolitical tensions on currency fluctuations — enabling the firm to hedge positions that saved millions in potential losses. This is not hypothetical — it is live risk management in action at one of the world’s largest financial institutions.

At the retail level, platforms like Riskalyze use AI-powered risk profiling tools that assign investors a numeric Risk Number on a scale from 1 to 99. This score quantifies personal risk tolerance with real precision and then aligns portfolio construction to match it — replacing the vague “moderate risk” checkbox that financial advisors have used for decades with something actually measurable. Monte Carlo simulations and scenario testing then model how the portfolio would behave across thousands of possible market futures.

According to IDC, 92% of organizations that have adopted AI tools report significant productivity gains, with average returns on investment reaching 3.7 times initial cost — and some organizations reporting returns of up to tenfold.

Understanding the ethical dimensions of AI risk modeling is just as important as understanding the technical ones. For a broader look at the risks and responsibilities that come with AI adoption, the overview of ethical AI trends every user should know in 2026 provides essential context for any investor using these tools.

5. AI-Driven Portfolio Optimization: Smarter Allocation at Every Level

Portfolio optimization — constructing the mix of assets that maximizes return for a given level of risk — is one of the most mathematically complex problems in finance. Harry Markowitz won a Nobel Prize for formalizing it in 1952. AI has now taken that theory and applied it in ways that were computationally impossible until recently.

Modern AI systems use machine learning to go beyond static correlation matrices. They capture non-linear relationships between assets, adapt to regime changes in the market, and rebalance allocations in real time as conditions shift. With global assets under management expected to reach $145.4 trillion, the need for intelligent, scalable portfolio management has never been greater.

AI portfolio systems identify over-concentration, flag hidden correlations between seemingly unrelated holdings, and recommend allocation adjustments before a human manager would even notice the drift. They also optimize for tax efficiency — executing tax-loss harvesting automatically when opportunities arise, improving after-tax returns without any additional decision-making burden on the investor.

Robo-Advisors: AI Portfolio Management for Everyone

Perhaps the most consequential democratization AI has delivered to finance is the robo-advisor. These platforms use machine learning to provide institutional-grade portfolio management at consumer prices — often charging 0.25% to 0.50% annually compared to the 1% or more that traditional advisors typically command.

Betterment, Wealthfront, and similar platforms automatically construct diversified portfolios based on an investor’s stated goals and risk tolerance, then continuously rebalance them and harvest tax losses without any intervention required. They bring discipline that purely human investors often cannot maintain during volatile markets — they do not panic sell in March 2020 or chase momentum in late 2021.

Magnifi takes a different approach — functioning as a search-driven AI investment assistant. Investors can ask plain-language questions like “Am I diversified?” or “Find me ESG-focused ETFs with low expense ratios,” and receive data-backed answers drawn from their actual linked brokerage accounts. It surfaces hidden overlaps, identifies concentration risks, and suggests actionable steps — all through a conversational interface.

HSBC’s Smart Advisor represents the institutional version of this model. The AI-powered system now provides personalized investment guidance to over two million clients, optimizing their portfolios dynamically based on both changing market conditions and each client’s evolving financial profile.

For investors who want to build the skills to evaluate and use these platforms effectively, the resource on the best AI platforms for online skill building covers the learning tools that can accelerate that process significantly.

6. Behavioral Finance Meets AI: Taking Emotion Out of the Equation

Even sophisticated investors are human. And humans, when their money is on the line, make predictably irrational decisions. Loss aversion causes investors to hold losing positions too long. FOMO drives them into assets at market peaks. Recency bias makes last month’s winner look like next month’s opportunity. Overconfidence leads to under-diversification.

Behavioral finance has documented these biases for decades. AI is the first tool that can systematically counteract them at scale.

AI investment platforms can be configured to enforce rule-based discipline — only executing trades that meet predefined criteria, flagging when a proposed action deviates from a stated investment policy, and surfacing data that challenges a human’s instinctive conclusions. Some platforms track the emotional content of market commentary and news flow, publishing a real-time Fear & Greed index that gives investors self-awareness about when the market — and potentially they — are being driven by sentiment rather than fundamentals.

Predictive Investor Behavior Modeling

Beyond individual behavior, AI can model how crowds of investors are likely to behave under specific conditions — making it possible to identify contrarian entry points before the crowd reverses course. When sentiment indicators reach extreme levels, historical patterns suggest mean reversion — and AI systems can flag those moments with precision that no manual analyst could match across every sector simultaneously.

This behavioral edge compounds the technical advantages of AI. It is not just that AI makes better predictions — it is that AI makes them while remaining immune to the psychological pressures that cause human predictions to deteriorate under stress.

7. Top AI Tools and Platforms for Investors in 2026

The AI investing landscape has matured rapidly. In 2026, the market includes everything from consumer robo-advisors to institutional-grade quantitative platforms — and the right tool depends entirely on how you invest.

For Active Traders

Tickeron is one of the most comprehensive AI trading ecosystems available to individual investors. It offers pattern recognition algorithms, signal generation, AI trading bots with configurable risk parameters, and portfolio optimization tools that combine Modern Portfolio Theory with machine learning. Each bot includes maximum drawdown limits, position sizing logic, and sector correlation analysis to prevent overexposure.

RockFlow’s Bobby represents a newer generation of AI trading tool — a full-stack trading agent that generates quant strategies, analyzes market structure, and can execute trades in real time. It is built for investors who want algorithmic-level sophistication without writing a line of code.

For Long-Term and Passive Investors

The established robo-advisors — Wealthfront, Betterment, and Ellevest — remain the most accessible entry points for passive AI-driven portfolio management. All offer automated rebalancing, tax-loss harvesting, and goal-based investment pathways. Fees are low and minimums have largely been eliminated, making them genuinely accessible to first-time investors.

Magnifi works well for investors who already have brokerage accounts and want AI analysis layered on top of their existing holdings, without moving assets to a new platform.

For Research and Institutional Analysis

At the professional end, platforms like Bloomberg Intelligence AI and Morningstar’s AI-powered analytics suite provide deep fundamental analysis, sector screening, and custom report generation at institutional standards. Jump.ai adds another layer — using sentiment analysis and semantic parsing on advisor meeting transcripts to surface insights, generate scorecards, and improve client communication quality.

Beyond investing platforms specifically, AI-powered productivity tools are becoming an essential part of the modern investor’s workflow. Managing deal flow, responding to investor queries, and staying on top of communications at scale is itself a challenge — which is why tools like an AI email reply generator can save significant time for anyone in a research or advisory role who needs to maintain consistent, professional communication without sacrificing hours to the inbox.

For a broader view of AI-powered email efficiency in professional workflows, the guide on AI email assistants that save hours per week is worth a read.

8. Real-World Use Cases: How Institutions Are Investing Smarter with AI

Abstract descriptions of AI capabilities only go so far. The most convincing evidence of AI’s impact on investment decisions is in the actual results being produced by institutions that have already deployed these systems at scale.

Goldman Sachs has integrated AI across its trading and risk operations. The firm’s Quantum Studio has demonstrated bond risk minimization of up to 40% through quantum-enhanced optimization — and its Risk Guardian AI model has been used to predict and hedge against geopolitical currency risk in real time. Goldman’s AI-powered systems have also accelerated M&A analysis, compressing weeks of due diligence into days.

Morgan Stanley has embedded AI into its wealth management division, helping advisors generate personalized portfolio insights for their clients at scale. Morgan Stanley Research’s AI exposure mapping tracked over 3,600 stocks to identify those with genuine AI revenue exposure — versus those merely mentioning AI in earnings calls. That distinction turned out to be worth significant alpha as the market began differentiating between AI beneficiaries and AI hype in 2025.

HSBC‘s Smart Advisor serves over two million retail clients with AI-personalized portfolio management — a feat that would have required thousands of additional human advisors under the traditional model. It has made sophisticated investment guidance economically viable at a customer segment where it was previously unprofitable to provide.

In the venture capital world, AI is reshaping how LPs and fund managers evaluate deals. According to the OECD’s 2026 AI VC report, AI firms attracted 61% of all global venture capital in 2025 — $258.7 billion of a $427.1 billion total — reflecting both investment in AI as a sector and AI’s role in identifying the deals. Sophisticated VC funds now use machine learning to screen thousands of companies against historical success patterns, reducing the deal-sourcing surface area before human judgment is applied.

9. Benefits of Using AI for Investment Decisions

The practical advantages of AI in investing are substantial and cumulative. Each benefit compounds the others:

Speed and scale: AI processes millions of data points simultaneously and executes decisions in microseconds — operating across global markets 24 hours a day without fatigue or attention lapses.
Analytical depth: Machine learning models now analyze hundreds of variables simultaneously. Traditional quantitative models analyzed tens. The additional variables often capture non-linear market dynamics that linear models miss entirely.
Bias elimination: AI does not panic, chase momentum, or rationalize a bad position because selling it would feel like admitting a mistake. It executes the logic of the strategy, consistently, regardless of market conditions.
Democratized access: Robo-advisors have made institutional-quality portfolio management accessible to investors with no minimum account size. The tools that once required a $10 million account are now available to anyone with a smartphone.
Continuous optimization: AI portfolios do not drift until the next quarterly review. They are continuously optimized — rebalanced, tax-harvested, and adjusted in response to changing market conditions without requiring any action from the investor.
Early risk detection: Stress testing and early warning systems that would have taken days to run can now be executed in real time — giving investors a chance to act before risk materializes rather than after.

10. Limitations and Risks of AI in Investing: What You Must Know

Balanced judgment requires acknowledging what AI cannot do — and where it makes things worse. Anyone selling a fully autonomous AI investment system as risk-free is either uninformed or dishonest. Here are the genuine limitations every investor should understand:

The black box problem: Many AI models, particularly deep learning systems, produce outputs that cannot be easily explained in human terms. When an algorithm declines a loan or triggers a trade, understanding exactly why requires transparency the system often cannot provide. Regulators and risk managers find this opacity deeply uncomfortable — and rightly so.
Overfitting on historical data: An AI model trained extensively on historical market data may perform brilliantly in backtests but fail completely when markets enter regimes the training data never included. The COVID-19 crash, the DeepSeek shock, and other black swan events demonstrate how quickly AI systems can be caught off-guard by genuinely novel conditions.
Data quality and bias: AI systems are only as good as the data they train on. Historical financial data contains systemic biases — it reflects the world as it was, not necessarily as it is. A model trained on decades of low-interest-rate environments will not inherently understand an inflationary shock.
Systemic amplification risk: When many institutions use similar AI systems trained on similar data, their models can produce correlated behavior — amplifying rather than dampening market moves. The 2025 flash crash incident, where a 6% S&P 500 drop occurred in minutes before circuit breakers engaged, is a direct example of this risk.
Information asymmetry: The IMF has flagged that institutional access to exclusive, proprietary data sets gives large players an AI advantage that retail investors simply cannot match — creating a fairness concern that regulators have not yet adequately addressed.
Regulatory uncertainty: The SEC, ESMA, and MAS are all developing frameworks for AI-driven investment advisory services, but those frameworks are still evolving. Investors relying heavily on AI tools should understand they are operating in an environment where the rules are still being written.

The most dangerous AI investment tool is one you trust completely without understanding its limits. Used as a co-pilot, AI is transformative. Used as an autopilot, it is a liability.

11. The Future of AI in Investment Decisions: Trends Defining 2026 and Beyond

The current moment is still early. The AI capabilities transforming financial markets today represent only a fraction of what is coming — and the trajectory is steep.

Agentic AI: Fully Autonomous Investment Management

The next evolution beyond today’s AI tools is agentic AI — systems that do not just analyze and recommend, but autonomously plan, execute, and adapt entire investment workflows without human prompting. Think of it as the difference between a calculator and a CFO. Early versions of agentic finance AI are already in deployment at hedge funds, and the capabilities are expanding rapidly.

Quantum-AI Convergence

Quantum computing will multiply the optimization power available to AI trading systems by orders of magnitude. Goldman Sachs’ Quantum Studio demonstrations are already showing what becomes possible — portfolios optimized across dimensions of risk that classical computing cannot even represent cleanly. As quantum hardware matures, the systems that integrate quantum methods earliest will have a structural advantage in strategies involving complex optimization, such as fixed income arbitrage and multi-factor equity models.

Explainable AI (XAI): Transparency Becomes Mandatory

As regulators catch up with technology, explainable AI — systems that can produce human-readable justifications for their outputs — will shift from optional to required in regulated advisory contexts. This creates a design imperative: the best AI investment tools will not just produce good recommendations, they will produce auditable reasoning that investors, advisors, and regulators can evaluate and challenge.

ESG Meets AI: Real-Time Sustainability Scoring

AI is enabling a new generation of ESG investing tools that go beyond self-reported corporate disclosures. Satellite imagery, supply chain data, and NLP analysis of regulatory filings can now produce real-time sustainability scores that reflect actual company behavior rather than polished sustainability reports. For investors with genuine ESG mandates, this is transformative — finally giving ESG integration the data quality it has always lacked.

Continued Democratization

Gartner projects that more than 80% of enterprises will have adopted generative AI tools by 2026. In parallel, consumer-facing AI investing tools are moving from basic robo-advisors to sophisticated personalized analytics. The gap between what institutional investors can do with AI and what retail investors can access is narrowing — not closing, but narrowing meaningfully.

Morgan Stanley estimates that nearly $2.9 trillion in global data center investment will flow through the economy by 2028 to support AI infrastructure — representing one of the largest capital allocation cycles in history and its own set of investment opportunities for those watching the sector carefully.

Staying current with these shifts requires continuous learning. For those looking to build AI fluency alongside their investment knowledge, the curated list of best AI platforms for online skill building is a useful resource for structured, practical AI education.

12. How to Start Using AI to Improve Your Own Investment Decisions

Understanding AI investing in theory is one thing. Putting it to work in your own portfolio requires a more deliberate approach. Here is a practical framework:

Step 1 — Define your goals and risk tolerance before touching any tool. AI platforms optimize for what you tell them to optimize for. If you have not articulated your actual investment goals clearly, the algorithm will optimize for the wrong objective.

Step 2 — Match the tool to your investment style. Passive, long-term investors get the most value from robo-advisors. Active traders benefit from AI signal platforms and algorithmic tools like Tickeron. Researchers and advisors benefit from AI-powered analysis suites. Using a high-frequency trading tool when you are a buy-and-hold investor is like using a racing engine in city traffic.

Step 3 — Treat AI as a co-pilot, not an autopilot. The most sophisticated institutions in the world use AI to inform and improve human judgment — not to replace it entirely. AI does not understand your personal financial situation, your tax circumstances, or the life goals behind your investment targets. You do.

Step 4 — Build your AI literacy. Before trusting any AI system with your capital, understand what it does, what data it trains on, and how it handles scenarios outside its training distribution. The investors who get burned by AI tools are usually those who trusted outputs they did not understand.

Conclusion: AI Is the Most Powerful Tool in Investing — Use It Wisely

AI investment decisions are no longer the exclusive domain of hedge funds and Wall Street giants. The same capabilities that Goldman Sachs uses to manage billions in risk are now accessible — in scaled-down forms — to any investor with a smartphone and a brokerage account.

What AI offers is genuinely transformative: the ability to process more information, faster, with less emotional interference, across more scenarios than any human team could match. The institutions that understand and deploy these capabilities well are pulling ahead. The investors who ignore them are falling behind.

But the technology has real limits. Overfitting, opacity, systemic risk, and regulatory uncertainty are not minor footnotes — they are genuine constraints that every investor needs to factor into how much trust they place in algorithmic outputs. The winning approach in 2026 is not fully manual or fully automated. It is a thoughtful collaboration: AI doing what it does best at speed and scale, human judgment applying context, values, and adaptability that no algorithm has yet learned to replicate.

If you want to explore further, a deep dive into the broader AI analytics landscape — beyond just investing — offers valuable context: the guide to best AI analytics tools for small businesses in 2026 shows how the same AI principles driving investment decisions are reshaping financial analysis across the business world.

Frequently Asked Questions

How can AI improve investment decisions?

AI improves investment decisions by processing vastly more data than any human analyst, eliminating emotional bias, identifying non-linear patterns in market behavior, and executing strategies with consistent precision. Specific applications include stock market prediction, real-time risk assessment, portfolio optimization, and automated rebalancing — all operating continuously, without fatigue or psychological interference.

Is AI better than a human financial advisor for investing?

AI outperforms humans in data processing speed, pattern recognition, and emotional consistency. Human advisors outperform AI in understanding personal context, navigating unprecedented scenarios, and applying ethical judgment to complex life situations. The most effective investment approach in 2026 combines both: AI for analytical horsepower, humans for context and accountability.

What AI tools do professional investors use?

Institutional investors use tools including Goldman Sachs’ Quantum Studio and Risk Guardian AI, Bloomberg Intelligence AI, and Morningstar’s analytics suite for research and risk management. Retail and semi-professional investors use platforms like Tickeron, RockFlow, Magnifi, Wealthfront, and Betterment for portfolio management, trading signals, and automated rebalancing.

Can AI predict the stock market accurately?

AI has improved market forecasting accuracy to approximately 90% in controlled backtesting environments — up from roughly 80% with earlier models. However, accuracy degrades significantly during black swan events or market regimes outside the training data. AI is a powerful probabilistic tool, not a crystal ball, and should be used to improve decision quality rather than eliminate uncertainty entirely.

What are the risks of using AI for investing?

The primary risks include: the black box problem (opaque decision-making), overfitting on historical data (failure during novel events), correlated algorithmic behavior that can amplify market volatility, data quality and bias issues, and the regulatory uncertainty still surrounding AI-driven advisory services. Using AI tools without understanding these constraints is itself a significant risk.

Are AI investing platforms suitable for beginner investors?

Yes — particularly robo-advisors like Betterment and Wealthfront, which require no investment knowledge to operate. They automatically construct diversified portfolios, rebalance on schedule, and handle tax optimization without any input from the investor. For beginners, they offer a disciplined, low-cost introduction to investing that avoids the most common behavioral mistakes.