IB Maths AI HL takes the five topics of AI SL and extends every one of them, adding 90 hours to reach the IB's recommended 240. Read topic by topic, the extra content can look like a random list: eigenvalues next to Poisson, graph theory next to volumes of revolution. Read as modelling, it falls into four strands: change over time, data and inference, networks and states, and space and motion.
This guide uses those four strands, with the official hours and papers, and a checklist for each. It follows the current guide (first assessment 2021), used for exams up to November 2028. If you want help working through it, our IB Maths AI HL tutors teach every strand below.
The official hours, SL against HL
| Topic | AI SL | AI HL | Extra at HL | Main HL additions |
|---|---|---|---|---|
| 1 Number and algebra | 16 | 29 | +13 | log laws, complex numbers, matrices, eigenvalues |
| 2 Functions | 31 | 42 | +11 | logistic and other models, linearising data |
| 3 Geometry and trigonometry | 18 | 46 | +28 | radians, vectors, matrix transformations, graph theory |
| 4 Statistics and probability | 36 | 52 | +16 | Poisson, confidence intervals, further tests, Markov chains |
| 5 Calculus | 19 | 41 | +22 | further rules, volumes, kinematics, differential equations |
| Toolkit and exploration | 30 | 30 | 0 | — |
| Total | 150 | 240 | +90 |
Geometry and trigonometry grows the most, from 18 to 46 hours, and statistics and probability ends as the largest topic at 52. The guide says all topics are compulsory. The four strands below cut across these numbered topics, the way AI exam questions do.
How AI HL is examined
Every component needs technology. Paper 1 is 120 minutes, 110 marks and 30% of the grade. Paper 2 is also 120 minutes, 110 marks and 30%. Paper 3 is 60 minutes and 55 marks, worth 20%, with two compulsory extended-response problem-solving questions. The exploration, marked by your teacher and moderated by the IB, is the last 20%. A GDC is required in all three papers.
Strand 1: change over time
Calculus and differential equations, plus the function families used to describe growth. This strand is where AI HL calculus goes well beyond SL.
- ☐ Derivatives of sin x, cos x, tan x, e^x, ln x and x^n for rational n
- ☐ Chain, product and quotient rules; related rates of change
- ☐ The second derivative, maximum and minimum tests and points of inflexion
- ☐ Integrating x^n, sin x, cos x, 1/x and e^x; integration by inspection or substitution
- ☐ Areas between a curve and either axis; volumes of revolution about either axis
- ☐ Kinematics: displacement, velocity, acceleration and total distance travelled
- ☐ Setting up a differential equation from a context and separating variables
- ☐ Slope fields
- ☐ Euler's method, for single equations and coupled systems
- ☐ Phase portraits of coupled linear systems, read from eigenvalues
- ☐ Second-order differential equations solved numerically as a coupled system
- ☐ Logistic, natural logarithm and piecewise models; composite and inverse functions in context
Strand 2: data and inference
AI HL turns AI SL statistics into a full inference course: from collecting data well to judging how far a result can be trusted.
- ☐ Designing data collection; reliability and validity
- ☐ Grouping data and choosing degrees of freedom for a chi-squared test
- ☐ Non-linear regression, the sum of squared residuals and R²
- ☐ Linearising data with logarithms; log-log and semi-log graphs
- ☐ Linear combinations of random variables; unbiased estimates of mean and variance
- ☐ The central limit theorem
- ☐ Confidence intervals for a mean, using the normal and t-distributions
- ☐ The Poisson distribution
- ☐ Critical values and critical regions
- ☐ Tests for a normal mean, a binomial proportion and a Poisson mean
- ☐ Testing whether the population correlation coefficient is zero
- ☐ Type I and type II errors and their probabilities
Strand 3: networks and states
Matrices sit at the centre of this strand. They describe a network, a set of states, or a system of equations.
- ☐ Matrix algebra, determinants and inverses; solving linear systems with matrices
- ☐ Eigenvalues and eigenvectors of 2 × 2 matrices; diagonalisation and powers
- ☐ Graph vocabulary: simple, complete, weighted and directed graphs; trees
- ☐ Adjacency matrices and the number of walks of length k
- ☐ Transition matrices built from a graph
- ☐ Eulerian and Hamiltonian trails, paths and cycles
- ☐ Minimum spanning trees with Kruskal's and Prim's algorithms
- ☐ The Chinese postman problem; travelling salesman upper and lower bounds
- ☐ Markov chains: initial states, steady states and long-term probabilities
Strand 4: space and motion
Geometry at AI HL moves from triangles to coordinates, vectors and transformations.
- ☐ Radians, arc length and sector area; the unit circle and the Pythagorean identity
- ☐ Matrix transformations: reflections, stretches, enlargements, rotations and translations, and compositions of them
- ☐ The determinant as the area scale factor of a transformation
- ☐ Vectors: components, scalar product and vector product
- ☐ Vector equations of lines; motion with constant velocity
- ☐ Motion with variable velocity in two dimensions
- ☐ Complex numbers in Cartesian, polar and exponential forms, and their arithmetic
Smaller additions elsewhere: the laws of logarithms, rational exponents, the sum of an infinite geometric series, and transformations of graphs. These rarely make a whole question on their own, but they turn up inside the strands above.
Where AI HL students lose marks
| Sticking point | Strand | What helps |
|---|---|---|
| Choosing the wrong test or wrong tail | Data and inference | Write H₀ and H₁ in words before touching the calculator |
| Conclusions with no context | Data and inference | End with one sentence about the situation in the question |
| Eigenvector and steady-state arithmetic | Networks and states | Check that the probabilities add to 1 |
| Algorithms written as a list of answers | Networks and states | Show the order edges were chosen in, as a table |
| Step-size errors in Euler's method | Change over time | Set out each step in a table with x, y and gradient |
| Mixing up rotation and reflection matrices | Space and motion | Test the matrix on the point (1, 0) |
Paper 3 and the exploration
In our tutors' experience, Paper 3 questions start with familiar steps and then extend an idea, often a model, into territory the student has not met in that form. An hour for 55 marks leaves little time to recover from a slow start. The exploration follows the same rules as at SL: 20 marks, roughly 12 to 20 double-spaced pages, and the work must be the student's own.
What changes for students who start DP in August 2027
The new AI course is first taught in August 2027 and first assessed in May 2029. The current course is last examined in November 2028. For AI HL the IB has published these changes:
- Removed: complex numbers, the vector product, the Poisson distribution and some hypothesis tests.
- Marks: Papers 1 and 2 become 100 marks each; Paper 3 becomes 50 marks in one hour. Technology is still required throughout.
- The IA: 30 hours, still 20%, with criteria for problem specification (4), abstraction (6), computation (4) and interpretation (6).
The IB says the new AI course has "a specific focus on mathematical modelling and statistical analysis".
What parents can check each term
- Which of the four strands has the class covered, and which feels least secure?
- Does your child write hypotheses and conclusions in full sentences?
- Have they practised a whole 120-minute paper, and a Paper 3 style question?
- Can they explain what their GDC output means, not just read it off?
- Is the exploration plan their own idea, with maths they can explain at HL level?
Our AI HL guides, by strand
- Change over time: AI HL calculus beyond the SL rules, slope fields, Euler and coupled differential equations, logistic growth and log-log linearisation.
- Data and inference: confidence intervals and AI HL hypothesis tests, Poisson distribution questions.
- Networks and states: spanning trees, Chinese postman and travelling salesman, eigenvalues, transition matrices and Markov chains.
- Space and motion: rotations, reflections and stretches as matrices, vectors and complex numbers for AI HL.
How Maths Bodhi helps
In the free demo, we give one question from each of two strands to see where the gap is, then match a tutor who teaches AI HL. Progress is tracked through written method, past-paper style practice and the strand checklists. For students aiming high who want a senior tutor, there are senior one-to-one AI HL lessons with Ajay Vatsyayan.
Home tuition is available in Gurgaon (Gurugram). Students elsewhere in India and abroad learn with us online. You can book a free demo class.
Written and checked by the Maths Bodhi team. Facts checked against official IB sources, November 2026.
Frequently asked questions
- What topics are in IB maths AI HL?
- The same five topics as AI SL (number and algebra, functions, geometry and trigonometry, statistics and probability, calculus), each extended, with additions such as matrices, graph theory, differential equations, Poisson and confidence intervals.
- How many teaching hours is IB maths AI HL?
- The IB recommends 240 hours: 29 for number and algebra, 42 for functions, 46 for geometry and trigonometry, 52 for statistics and probability, 41 for calculus and 30 for the toolkit and exploration.
- Does IB math AI HL have calculus?
- Yes, 41 suggested hours, including further differentiation and integration, volumes of revolution, kinematics, slope fields, Euler's method and coupled differential equations.
- Are all IB maths AI HL papers calculator papers?
- Yes. A GDC is required in Paper 1, Paper 2 and Paper 3.
- Is the Poisson distribution still in AI HL?
- Yes, for exams up to November 2028. The IB has removed it from the new AI HL course first examined in May 2029, along with complex numbers, the vector product and some hypothesis tests.
- What is the biggest topic in IB maths AI HL?
- Statistics and probability, with 52 of the 240 suggested hours. Geometry and trigonometry grows the most from SL, from 18 to 46 hours.
